Tonmeister Technology Recording Environments Sound Sources and Microphone Techniques 096172000x 9780961720001 Compress

MICHAEL DICKREITER
TONMEIS
sunt
i:
TECHNOLO
RECORDING ENVIRONMENTS, SOUND SOURCES A
TERPRISES,
INC.
Dickreiter
Tonmeister Technology
Digitized by the Internet Archive
in 2022 with funding from
Kahle/Austin Foundation
https://archive.org/details/tonmeistertechnoOO00Odick
Michael Dickreiter
TONMEISTER
TECHNOLOGY
Recording Environment
Sound Sources
Microphone Techniques
with 157 illustrations
and tables
Translated from the

German by
Stephen F. Temmer
Temmer Enterprises Inc. 1989

German edition issued under authority of
The German Broadcasting Technical Training Center
Nuremberg, F. R. Germany
Title of the original German edition:

"Mikrofon-Aufnahmetechnik. Aufnahmeraum
Schallquellen, Mikrofon-Aufnahme"
© 1984 S. Hirzel Verlag, Stuttgart, F.R. Germany
© 1989 Temmer Enterprises, Inc.

767 Greenwich Street, New York, NY 10014
TEL 212-741-7418 FAX 212-727-3870
All rights reserved. No part of this publication may be trans-
lated, reproduced, stored in a retrieval system, or transmitted,
in any form or by any means, electronic, mechanical, recording
or otherwise, without written permission from the publisher.
Printed in the United States of America
ISBN 0-9617200-0-X
Author’s Preface
There are two readily definable areas of responsibility for the performers and
the studio engineers between the musicians in the studio on the one hand and the
loudspeakers in the control room on the other, and even between an announcer in an
announce booth and the headphones worn by the sound engineer: one is the sound
environment (usually the studio) with its sound sources and microphones, while the
other is the control room with its numerous technical installations.
The meanings which these areas represent are very much dependent on the task
at hand. For serious music it is the studio and the microphone setup Wvhich play the
most important role,while for popular music the microphone setup is important but
the recording can never be a success without a very comprehensive technical arsenal
in the control room. This book deals mainly with the area of(sound studio recording)
which is often referred to as(microphone technique.) All this is only meaningful if the
sound environment and the sound sources are included in the discussion.
The areas of recording environment, Sound source, and microphone technique

are arranged by key words according to subject matter. Their relationship to practice,
the day to day application of these facts, is placed in the foreground as much as
possible. But since taste and the desired resulting sound, i.e., subjective judgments,
play important roles, it is very difficult to pass on specific recommendations.
Therefore, this book restricts itself to that knowledge which forms the basic
prerequisite for achieving any desired sound. One may only build one’s personal style
on such fundamental knowledge.
It is the author’s hope that the information contained in this book and the
prerequisite knowledge will be of assistance to the neophyte as well as the seasoned
professional. This book’s main thrust is professional studio technology. The acoustic
parameters as well as the microphones employed are not solely high-end professional.
Therefore this knowledge is applicable to non or semi-professionals as well. I am
especially indebted to Karl Filbig and Norbert Kloevekorn who read the manuscript
and enriched it with suggestions and improvements from their own professional
practice.
Nuremberg, July 1983 Michael Dickreiter

Translator’s Preface
What are Tonmeisters?

Tonmeisters are persons responsible for those sound recordings and transmis-
sions which have artistic content. In order to fulfill both their assigned and self-
generated artistic intentions, they apply means which go far beyond the purely
technical recording or transmission requirements. Their functions include consulting
with the artists, personally influencing their interpretation, and applying their own
creativity to the final product. The tonmeister may be found working in the areas of
radio, television, film, sound recording, and theater sound reenforcement.
The tonmeister strives to achieve a significant balance between the various sound
sources through his careful selection and meaningful application of microphones and
other technical devices. During the recording or transmission he supervises the proper
match between the intentions prior to the task and the performance at hand, and
decides when corrections or retakes are required. After the event it is he who selects
from among the numerous takes and combines the component parts or tracks into a
meaningful final product.
The performers, artists, conductors, composers, authors, and directors must find
in the tonmeister an artistic partner whose judgement and critique are above
reproach, and who is a determining component in the realization of their intentions
and the full unfolding or their abilities.
It is a precondition for the success of tonmeisters that they(have the ability

critically to listen, \combined with fan above average, psychologically healthy hearing
apparatus, |a(creative ability for organization,\ as well as\knowledge of the problems
of artistic production} They must also be aware of the connection between the artistic
and technical possibilities, and they are therefore required to possess an artis-
tic/technical dual talent.
The reasons for this book

This book represents a first in English language publications: a text which totally
integrates music and the technology of musical recording and transmission. The
tonmeister concept was born in central Europe virtually simultaneously with the
advent of recording and broadcasting, and it is gratifying to this translator to see an
ever greater acceptance of the concept in the English speaking world as evidenced by
an increasing offering of tonmeister-like courses at institutions of higher learning, and
an ever greater degree of acceptance of graduates of such programs in industry.
This text is ideally suited to the short-course seminar of two to six weeks, but
may be used as a component text in regular undergraduate and general studies
courses as well. It will also be found helpful to professionals in the industry who will
find in its pages a refreshingly direct discussion of the basic principles of musical
engineering which may have faded too far into people’s distant past. It will also serve
to stimulate interest in the tonmeister program among young people interested in the
recording and broadcasting fields. It is towards this end that the translator has
engaged in his translating effort.
TABLE OF CONTENTS
EL ROUEN TEV LERCOINS

INL Pe Sod aig san a tnaly oe net eee t rtiae aoe
DOUNC INaVES INCAINGMOGIN: ay ccda GY CS Gti wionhe es ta ne he ee ere
i SOUP APE ent viene st Se Solar laaty oily atten ior
! SOUR IWEke PLODCIUCS eed cs ce ete eee eee tas
13 SOU ata URICTE Oe eee, galt Ae ba ch SC eee anes tet ee,
OOMMACOUSHCSM UTC Aine ttc Lone aeesient aeirat) Steere) ountee gtemmnnueny. yeeti
Influence of acoustical space on the sound event .............
EUNUAMCN Cas Ol LOOM BCOUSIICS —AaNty ss sy. orhee es care wey eee
PUG iiais Or AUlAl ACOUNEICS a Sipe sn ete tr Pe es sete tore eH
NDN
Significance of room tone for microphone placement and the
Stemi eNDerIenCe. any ey Whey riety em oe eet os ee
PIPER SOU Oo eetce Ber ee als orl Ped nits soso en emer nn mes
CseOimmedio-PiopALaLiOluttemuatiOn! "7 eee. ce ae eee as
Influences of temperature, humidity and wind ................
OUNCE AHUrActiOn ArOUNg OUStACIES cms sn). 1 aie de ss os en eo
Sound absorption for sound radiation over an audience
(CEL ELEVagBue: MGI mai nrgp a art a eel coeagechat be tea
The meaning of direct sound in récording’ “2257. 7.5 ty ee
DUNMOW AVO-TETCCUUNG inn etc ysays ae Siac ois ty oa Cee eear mer omer eay tens
CeCAOl Ol SOUNG TCHOCHOUS: 22 olz 0 v hud aed ele amare hare eae
FAICCL Ol SOUNL TEMEOCHONS! seagate ste meee cle ne earn
Wire and:talse sound source: localization” {,10 5 se, pee hate
Mncre ase In sound SOUrce-OWUNesstlCVel “Stas ay etm eee go
SHUM OLOL ALON © ond fons cite ire ea eters eee on tee Re ea
ADPpurent Toomsize-une spacial iiipressiON “fv ay op oe Oe
ECNOs Lute eCnO. TALI COMO” ciara ais tae otce tee ete oe eee eet
PROPER ren Re ree eet ieee ir tice atte oewoke ee he ee come oer mele rater
Piven trequency aDsOrDersd tio S a oe ee ies « fden aoe eee eeers ©
I AL CCDENCY SADSOLDEUS plat see here ne ri a Oe ee oe noe
DRIGEr ANSE ADSOLUCES: by tutors ans Gane he rge-e rae Ae © ole eee oear ctr a
AANA, Tete CUO TOONS: Onna, 01> core Fue r matin own te)x ius eae
Reve te AOU er Acre eae mete tres re Ser ee Fe Soke y ae erat
“7 2 tie DNC UOMICMOULIEVCLDC AOU Bence nds ho eer Oe er Neate oe
Ny sS Properties of PEVETDSIALION rte igatey Cyt ae eee oa ee
IRCVERDCIANION TaGtUGe ois oy vinta! siete = ita ie ae SB oe eae hee spate wlavo
Relationship between direct and diffuse sound ...............
Revyerberation radius corrections in’ practice “int... a2 are eo oo
Audible sensation and reverberation radius .................
SIND COIfos ae a all ihe aa ee onraiie oe ats & AB rn ars go rim era) GO
Orchestras and chamber music*ensembles “io 2. ee ee ee te ee
Tiig(rnientAcomiigiratwons (Aa titlcap acedele ay itecetae ie oe
BOA ATT ANC CIIGINS oe ecg tr oe hs sun nee hele fk ee riety ee
Dynamiceancesand level: 62,2 Svinte wipes 6 se nee qin anes oon baa hs
DiISHealeAvintrUEMCACOUSIICR” “Fores ae cite tee ne Se eee Eee 7
Cig A TLANY SIS WET CLONGs oH Sopra chebtide ay facie ov dum GlG-are ae) o>
Quasi steady-state oscillation characteristics ..................
Musical instrument dynamic range and loudness ..............+.4- 40
Technical dynamics: << s:cicete » econ ogo eae Cee ees 40
Level andoudimess® <2 sce oyu Sees is erecta ere arent 40
Leveltand timbre dynamics. ieee es eee ere oe 43
Strife MMStUNTENtS: “eens as pee i ee SoA ete ee roms ere rates 44
Instruments sa. 61s apd 4 & ahaee SURE Rd BP le veda eee Beene 44
Application 6 svi s!senie sted erasere + ofa abet ane teceaey Ruane da
SOUR ACOUBLICS <n 3/4 o-d-n hase sl208 eo °reeee ete eee tamer Sano ou 44
Radiating characteristics s:03. 5.0 Ge sonnei eeriecones 47
Woodwind inistrmemtss. ofa cau teaates sai auch cu eee oeceeet ne ore 48
TST DS » 4 (ene 4om, @ Soom amb ephedrine ee eae ect tee 48
FaNi,s liters lt) ee ee ee pee eee ene. Tete i) here 48
Sound -ACOUSHCS: a niktas, ox: ccakeoneennaing Mn navawee Ciesiy baa Retiome tag 48
Radiating charactertstic .-. t4.cP gcatae citys ates tila danayeane Pe ak sl
Brass “wind ANStrUmMe IS «dies carton sibaie celarecte > sakes enaaustap esteem amie 52
Instruments and. their application iipiugrapebencrens) eireephataee anmeoe~ 5 a9 32
Sound acOustles? av. G acd aa doaceeuaytes see ioc vant Reka See Bye
Radiating characteristic... «chaser oro tes ioleencuegouentg hitces arith kere 55
Percussion- Instruments <and\plane ¢ ecu. tutors: cianeites eie1 wien ee hie eins 56
Instruments and, their application st... aragcon scion bay-aittectn teense eecae 56
SOUMA ACOUSTICS 15 a5 ere sac buceitnne eare sie ait avnlerameiatesies ene nette, 56
SPC akin cand «SINC VOICE wn,tire,«ie bile kak 2 Gide Re ioeucitaatal. Saybeoben. cima 60
SDEAKING VOICE! «4 tawace Ble aaron tetas, aul Mis eae eons ee ae 60
SINGING VOICE. axis wanie eran ais, eee a ont ee eee ier eee 63
MICROPHONE. RECORDING ® oy cu coasts ipcaarnd ua Maen anee Wai meee wana teh 65
Microphone directional cchharacteristies iy c:cx.case saison cacti mae Se es 66
Receptor and transducer prniciple A ss.esueect tse diese a oe oe 66
Directional-characteristic. definition: a, os ein eos Sei 66
Omni-directional characteristic of pressure receptors ........... 66
Pigure=-8 Characteristic: a.= caer epee aiatnuseba seaa eee ee teed 69
Gardioidchanactenistice 2...24-siaiicuaite@ecieas: seaeheet Genk a; Oona 69
MIYper"and. Super-Cardioid «cine t ath wane nan pre Gusher aati Stunts 69
SWUCHADIS GIFECHONA DATIOTNS ain. adel shes cxueiannt onsie ass fat 69
Micropnomesfreqiency, FESPONSS., <ycns-die huaeenecwutaeertay Rireicikin Cursos 70
Directional characteristic frequency response. nsec os ete & on 5 70
Diréct.and diffuse field frequency response +. 40. 2. aagiassateoee 70
Dynamic andicapacitor microphones Ayuoser. aterm eientid ae bs a 73
PrOXiMAULY CITECts fic, act So's tay ean Sree Pun monet re igual eames ame 73
SPeCcial /PUrpPOSe MICTOPDORES oe <a 5-nsands om ooo sensi se eee cee 74
SOlOISE AMLCTO PROTOS caasesyan yh 5 oe psceease eateet ances eles ea a eae 74
Noise canceling mmiccomboness jvm tae house isn: ceietas ta berRiee chin en 74
Lavaliere micronnones ives. au.ctiine tenella ctonitomis dette enia uate 74
LinesOr Shot-gun. MUCTOPDOMES:s cncnercra) ue conteetantne: asst he +8)
Measurement micropnones Fis. atnicc cats ath ceases 77
Boundary surface: micropNones ncaa sitiecute eat erin ie ci eae a|
Spacial. MOAR Msc shaun. a:nota soi ancedemas amc I ieee fincas wate aa cn 78
SPacial PercepuGm. Of AcSOUNCSOUnCE as kee ceeds aeceeeae em 78
Distanpubeariny.” 4.5 0a austebustnds retentcelta Mariel ik ena emeh tian 81
Spacial hearing with loudspeaker reproduction ............... 81
DHELCOUINONY pate co uterca a og 5 URN at: OR RE eo ae 82
LOTCOLIMGNIOUS smear uctin aly ag gals wise Me Pe Paw Fe 82
Recording methods for room referenced stereophony .......... 82
Relationship between the L and R signals. ..... 006 seme es 85
PEELCO) Sites UROMILOTIID os garlic ciara aten Flee GK wince She ie: a oe SMT eh RE 86
PLOgrat level Melerse Ga sin wre sive op & waa CO eke 86
Correlation, coelicient meters: Weis tainea aed eee 86
DLETEQ SOONG, Got yon Bienes ine wx thas 5 exbanechs gaia alte de Sty ET 89
Loudspeaker monitonng requirements sera. eeate ee ov ier eaa - 89
MS -anG XY MICKOPHOME TECOMIGUES oo hoe cis, heer eee eee 90
Mednd Ss. : ANULY Signals... mathwa cravat ain boner ee 90
Sctiing up.d comcident microphone pairs) = cain wos aie eee 90
DUUSESOUNG. ein cnie. Sirntucs ine oce SECT RISO Wee SRE 93
MS microphone system practical advantages ................. 93
Multi-michophons technique =... 4. <<) sesieesiet shen oe G~ Saher ee eee 94
Principle: andjapplicationss4 na) aatie digi ere Sern ier, AoA 94
SGatny OF PDE INUSICNADS 65.4 cxu va)ih cette twat cern ae eter tie ee 94
ACOUSTICAL SOD ATAION sine asausnssices¥. a nik iste! ade RE Re OP ae 94
NOCTODUODe type ald SOUD cee a Ries ore tee. ese. 97
Individual microphone and overall levels. 1a... sae ome eo 2)
Pime-Or-atrival Stereo. techie (AB i eg asc ao come nonstate Mane cee 98
PING ple 2G. aD OMCatlOnle . «emcein iy teas araniiaat Leama Botta 98
INNCTODNONE VPS ANC SERIE 6 nus,cores 2 oeuecath Ge Aine. tet eee Re 98
IMIENE SUC CO LOCUINO MCS |hve to Satan sas voyyard faye. alle hs pa Na 100
PRINCI tN app UCAN ON dacs cnn Gis bug casuste vat, Greg eae 100
NUCTODNOTIC Lype. aldSCMIDs 5 io. nic) cuaneee ca peau eae 100
SHO DOl IUCTOP RON CS 9nates est a accehs eget Gea eae ie erate Adres ae 104
Improvement through the use of support microphones ........ 104
NGOTO SUpPOLt MICTQDNONES ac eye h teen, panda ein aaa 107
SEETCO SUPOOIt TNCLOPMOUES. moyrisiy rane ih Pati ea esmnin oe 107
Sound coloration; the small room sound ................4-. 107
Duimimy-néad Dinaural echmiQue.. 0. Gin e c ee ota Pa ee 108
|g1606)|0)LRN ar Noe inti aR aan aa en te Sars herPR@eent To) Bars 108
RE PTOCUCHON ear cones vustace aeon Gateuegetc tsculehWay ns ait oySe aT Tore 108
PNDDUCONOU: Mort cragasn oh wishes Sete Sesh is en are Ane MR eee ae 108
TRE TECOTIBG OL SIC Cons UUINC IES. gs ayy Wy a ueceee een cee ee gree 110
DirectionalacharactetStic eto Lin ait Sm uae ieee wean mea erie arr 110
PXIVeIMely -ClOSE TECOLUING a ctcmen ewes: vac hat asttnsinadaan eteee akoe 110
The Tecording OF wind TMStTUuIentS ss ccnas Moners e cisgs, san crenata ae at £2
DirecrOnaAl cNATACteniSUCs 4 oi. ah5 Ewe 2 asks felance) Pee eng teu yeneee 112
IWOOGWIIG MIStUNIDCIS | ifs ntact Sac a uanteeal oTua atte aca s ueece 112
Brass: WINK UIBEPUIMEINS os ooo cassie. a: en le ue AS ete ingot ales NSS
ENE TECOLdING OF DETCUSSION: INSITUMENES aon ce tesncgeie- te< cae eee he 116
DETIOUS MMSUC anc cscscea ete Mot ath cree. elon tee eee eee UG
Semt-classical and popular MUSIC coir, elt s ea ee te aos 116
Daz ADGTOMC MUSICS,» worachy: Micoeee Ss ie)Seeds ete na einus eA 119
Ae Tecordime Ol CUItaTs. ohio aoa tentalain ovlgeh soa eee anne eeaeing (5ene 120
Classicaleand acoustical guitars” Grawa asc oats oe ers sicitoe 120
ICCC CMU Piet li oon akc eet A ee ee ete sts, 120
The recording:of keyboard mstruments A225) «eee ee eae coe ee 122
PiatiO a Sn aha hie Oe, 4st a tek Bee alee se ee ee ee 122
Other: keyboard instruments "0 se fete i en ee eee 127
Keyboards is 5 sistig. santa aes eR SE ene, reel a aera 125
Pipe: Orga fe ivks 144 kkk Bee Ree © Wane eee one see amen 125
The recording of speech 9.4. 2hn eae nindck sen ee me meee eee 126
Playback loudness ‘and. frequency Tesponse 7.40 wees ts 126
AnhOUNCerD TECORGING: > :Agin seme aie tweak mse rae eee Lee 126
Interviewsyeye witiess-accounts elo... cen real Os na 129
Round itablewliscussions.. 12.0 swan. sb eeeae vies Ho ees oe 429
Fhe recording of-vocalsoloists and chorus, «caves ass hte ee 130
Vocal-soloists:(popular music)? Sieve aes cect ne ale Bae 130
Vocalsoloistsx(Sertous Music). nwa Rice neu aun eee a 130
CHOPS ests cts aise wane SO th ade ge ees Pe ne, 133
vesthetic priticiplesinimusical TEcording ) «<0 .ue vee ee oe 134
Distribution of sound sources along the stereo horizon ........ 134
Widthofithe-steréohorizon, ¢ 24.4. 4< hoki ho eee ee ee eee 135
Depth perspectives citi yiens 02.65. 5 hse 6 ye, eee A ee ae 135
Semi-classics, popular music, folk music, jazz ............... 135
TIUStrAtIONGSOULCES ey, cers ace a Sena Oh ee atts A ee eens 136
NTN AO Xs hes FOS RTS RA A oS orcs co:Bethe OSI ae Ee en ee ee 127
THE RECORDING ENVIRONMENT 1
Sound waves, fundamentals

Room acoustics, fundamentals
Direct sound
Sound wave reflections
Absorption
Reverberation
Reverberation radius
The acoustic phenomena surrounding our lives, in particular noises and speech,
and those within the province of art and entertainment rethe main music and speech,
are always perceived within an audible, acoustical space} This space may be small or
large, even infinitely large; it mayy be reverberant or dry. This environment serves the
same purpose in acoustical—especiallyinelectro- acoustical —— energy transmission, which
light serves in painting or sculpture. Without its environment or space we cannot
perceive acoustic phenomena, without light no optical ones. Space and light, therefore,
are sensory media. They are not neutral but rather change and interpret the occur-
rences. The acoustics of the recording space for music or speech adds information
about the nature of the room, about its dimensions and the nature of its walls. This
may well be information about the cultural or social environment: a highly reverberant
environment may give the impression of a church with its impressive silence and its
religious contemplation. There also is a considerable difference whether chamber
music is acoustically transmitted with the intended acoustics of a chamber or with that
of a 2000 seat auditorium. The proper acoustical space is just as important in an
historically proper recording as are the historical instruments. Because of the fact that
the optical impression is missing during electro-acoustic reproduction, the acoustical
spacial impression gains that much more importance.
Modern recording techniques for popular music, and to some extent even for
serious music, largely ignore the natural acoustical space and substitute a space illusion
generated using the electronic effects available in today’s control rooms. This does
not mean the ignoring of proper acoustics, but rather the optimizing of the space
illusion which otherwise would require great effort and time. The same measures also
permit the creation of artificial recording rooms, so that the space becomes part and
parcel of the esthetic task facing the tonmeister. Between the natural room acoustics
and the artificial room acoustics, which may well be impossible to imagine from the
purely physical standpoint, there are of course any number of transitional steps.
D The recording environment
Sound waves, fundamentals
Sound propagation
Sound travels in air as a lengthwise or longitudinal wave (fig-A). Zones of
greater density air with increased air pressure alternate with those of lesser density
with reduced air pressure, all traveling at the same speed of sound. The individual
particles of air only oscillate to and fro and, therefore, the sound wave does not
transport particles but energy: sound energy. This sort of density wave is the way in
which sound is transmitted through air, gases and liquids. In_ solids we also find
transverse waves (fig.A), aside from these density waves. Only in a propagation
medium which is virtually unlimited, do pure longitudinal and transverse waves exist.
For sound propagation in air, this room acoustics condition is adequately fulfilled.
Wave propagation in solids, on the other hand, usually travels in strings, plates, etc.,
ie., in limited media. Therefore waves propagating in solids generally have more
complex wave forms. For example, these are bending modes in plates and mem-
branes and expansion or torsional waves in strings and rods.
Since sound in air travels from the sound source evenly in all directions, the
sound waves may be compared to concentric spherical shells constantly increasing in
size and distance from the sound source, if it is small compared to the wave length.
Such a wave is described as a spherical wave. At a great distance from the sound
source the curvature of these spherical shells becomes so small that the spherical wave
gradually transitions into a plane wave (fig.B). In a spherical wave the amplitude of
the wave propagation decreases with increasing distance from the sound source, while
for a plane wave it remains theoretically constant, but in practice decreases but little.
In the immediate vicinity of the sound source, the sound wave has complex properties
(+p.73). In practice there are further influences on the sound propagation (+p.10).
Sound wave properties

Numerous physical magnitudes change during the propagation of a sound wave;
for example the location of the air particles in the room, the density of the air, the
sound pressure , e.g., the change of the barometric pressure resulting from the sound
wave, the sound velocity, e.g., the motional velocity of the air particles, the sound pres-
sure gradient, e.g., the pressure differential between two points in the room, and the
temperature. {Microphones react either to the sound pressure gradient or to the sound
pressure (capacitor and dynamic microphones) or to the sound velocity (ribbon
microphones)(~p.66).
The sound pressure is superimposed as alternating pressure on the constant
barometric pressure. The pressure variations are very small. At a distance of one
meter from an announcer they amount to approx. one millionth of the atmospheric
pressure. Sound pressure is equally effective in all directions. There being no prefer-
ential direction as might be indicated by the air particle motion. A microphone which
reacts to sound pressure variations therefore is equally sensitive in all directions
(omni-directional) if it is small compared to the wave length. The magnitude of the
sound pressure may be given in various units: normally by its rms value in pascal (Pa)
or newton per m* (N/m ); in previous times and now obsolete, in wbar (1 ubar= 0.1
N/m). In many cases the sound pressure will be given in decibels (dB) referred to
the standardized threshold of hearing (0 dB = 2 - 10°° N/m’)(fig.C).
Sound waves, fundamentals %)
I
bi A. Sound waves “1
Longitudinal wave (propagation in air
B. Sound waves in the

near and distant
sound field of a
sound source
Transverse wave
ily
Spherical wave in the near sound field
Plane wave in the distant sound field

4 The recording environment
frou Sede eae
C. Sound waves:
fundamental
mathematical
relationships
| | |
amplitude
| |
wave length A
Is. duration 1 bi|
MSC ud } = wave length (m)

Sic! c = speed of sound (m/s)
f T = duration, time (s)
fae f = frequency (Hz)
;
sound pressure levei (dB)

0 20 40 60 80 100 120
ie
2Oe ad id a oeOle 1 20
sound pressure (N/m: )(Pa)
For the plane wave: Q = air density (kg/m*)

ae p = sound pressure (N/m’) or (Pa)
Friis v= sound velocity (m/s)
ee Z = characteristic acoustical
=, impedance (Ns/m*)
time (ms)
D. Relationship 2TA 6 Se Ober 4G 18) 920) 2252455265 28s Seis
between
elapsed time
sie patesd OVA TZOR 2 128 AV ANEA CSS. 4NGGINGCif.008.256.8.9.0 | OlL2aem
istance for
the propaga- distance (m)
tion of a sound
wave in air
ea Sa ee 5
eo!
ad eae =
E. Relationships | $ Oo
of tone, fre- L mare: : |
quency and is con teas eS
wave length Oe a eo
forasound frequency (Hz)
weve: Na 33 65 131 262 523 1047 2093 4186 Hz
wave length
IOS 525. 2.6 les) 0.65 0.32 016 0.08 m
Sound waves, fundamentals 5
The sound velocity, the motional speed of the individual air particles, displays
different behavior from the sound pressure in the vicinity of most sound sources.
While the sound pressure increases without regard to wave length in direct proportion
as the distance from the sound source decreases, the sound velocity rises dispropor-
tionately.( Since this effect depends solely on the relationship between wave length and
distance from the sound source, }the effect becomes more pronounced towards low
frequencies and results in a bass boost in those microphones responsive to sound
velocity. mr iq
The sound pressure gradient, in other words the pressure difference between two
points in the sound field, behaves like the velocity in the near field. [Ail directional
microphones are pressure gradient transducers,Jexcept for certain types such as ribbon
microphones, which are velocity sensitive (+p.66). i
Sound pressure, sound velocity, and sound pressure gradient are in phase and
proportional to one another in a plane wave field. As one gets closer to common
sound sources, there is an increasing phase displacement of up to 90° between pressure
and velocity in the near field. Sound pressure and velocity are no longer proportional
to one another. The relationship of pressure to velocity is known as the characteristic
acoustical impedance (fig.C). In the near field it is a complex magnitude, while in the
distant sound field it is a physical constant of air, i.e., its acoustical resistance.
The sound speed refers to the sound wave’s travel through air in a room.) By
contrast to the sound’s velocity, its speed is a constant, solely dependent on atmos-
pheric conditions. Fig.D shows the relationship between distance and the time
required for sound to travel the same.
The wave lengths of sound waves which are of interest in sound recording stretch
from 10 meters (33 ft) for the lowest frequencies to about 2 cm (0.8") for the highest
frequencies (figs.D & E). Since the wave length or frequency of sound plays an
important role, they must be considered whenever dealing with sound reflection,
diffraction, radiation, absorption and the behavior of microphones in the sound field.
Sound analysis
The sound given off from natural sound sources is made up of many partial
vibrations. Sound analysis refers to the determination of frequency, amplitude, and
sometimes also the phase relationship of the various partial vibrations. Periodic sound
events consist only of harmonics (overtones, partials) and their frequencies are integer
multiples of a fundamental. Non-periodic sound events consist of an infinite number
of closely spaced harmonics. To this group belong noise and all unique events such
as transients. For natural sound sources, periodic and non-periodic sound events are
present simultaneously. [Speech and noise are predominantly non-periodic music and
tones predominantly periodic sound events.)
Specific analyzer methods are available depending on the task at hand and the
type of sound to be analyzed. For steady sounds there are the search frequency
method and the Fourier analysis, and for certain tasks swept filter or paralleled filter
circuits (third octave or octave analyzers). Variable sounds are analyzed mostly using
third octave and octave band analyzers. The real time frequency analyzer is ideally
suited for all sound events, since it presents a new spectrum as often as every 200 ms.
The device calculates the spectrum almost instantly from the digitized signal using
so-called fast Fourier transformation (FFT). From the rapidly changing instantaneous
measurements it is possible to calculate generally valid data, using statistical averag-
ing.
Influence of acoustical space on the sound event

Sound sources for recording generally are located in an enclosed room, not only
because of the practical advantages, but principally because the room changes and
augments the sound event in a way which the listener deems necessary and desirable.
If we partially or totally replace these room influences with artificial reverberation,
predominantly because of its controllability, subsequent addability or for commercial
reasons, then the same concepts apply. \
The room influences may be described in two ways:
1. Objectively through measurement of the sound events and their variation with
time (rogm acoustics),
Die‘Subjectively through verbal description of the audible experience (aural
acoustics).
™ Both methods are necessary; depending on the question at hand, either the
objective or subjective one may assume the greater importance. The objective con-
cepts of acoustics have been universally defined, while for the subjective acoustic
experience the concepts enumerated below have found wide acceptance. It is the aim,
but not the fully resolved problem, to unite the concepts of room acoustics with the
concepts of aural acoustics.
Fundamentals of room acoustics

When sound waves from a sound source spread within an acoustical enclosure,
they quickly meet up with walls, ceiling and floor and are reflected from there as sound
reflections (fig.A). These reflections meet up with these room boundaries again and
again, resulting in an increase in the density of these reflections with time, as they
arrive at a point within the room—at a microphone or listener. The first reflections
possibly may be heard as single reflections—as an echo—and be deemed disturbing.
They have a significant influence on the perception of the room and its influences on
the total sound picture, even though they are hardly ever heard (+p. 14). [The ensuing,
densely arranged reflections, together with the early reflections, form the meee
(+p.14). Since the sound waves lose energy with each reflection through absorption,
the intensity of the reflections, and with them the intensity of the reverberation,
weaken rapidly following the end of the sound event. Reverberation time is the
measure of this behavior. For a continuous sound event, an equilibrium is formed
between the added and absorbed sound energy in the room. This condition doesn’t
exist initially, but rather builds up during the reverberation build-up fig.B).
Sound emanating from a sound source reaches the listener or the microphone
first as direct sound (+p.10), and subsequently as the room reflections form, the diffuse
—
sgund (room tone). While the direct sound decreases rapidly withiincreasing distance
from the sound source, the diffuse sound is generally evenly distributed in the room
(fig.C). This means that the ratio of direct to diffuse sound varies with the distance
from the sound source, and this becomes an important criterium for the proper
microphone distance. At the reverberation radius (+p.26) about the sound source, the
direct and diffuse sound are present at equal levels (fig.C).
Fundamentals of aural acoustics

We use the following descriptive terms to describe the subjective properties of
the aural acoustics of a room: (fig.D):
A. Sound propagation
in an enclosed room
(the rays show the
propagation direc-
tion and intensity)
sound level
B. Direct sound and jdirect sound

diffuse sound (re-
verb build-up,
early reflections —— first reflections
and reverberation
decay) in an en-
closed space; time
study
reverberation
short duration sound event
sound pressure sound pressure

-_----4
i“
Faas
larger distance from the sound source smaller distance from

the sound source
CU) direct sound
—— diffuse sound
— meowol total sound

The recording environment
Sound level
eae radius
-— Total sound level

/
ne ee
= pase Mt KANCacFD SN (SW emtEa
ee re,
ce |
ek ee eee Ss oe aoe
+—->—-++ t+ + It
te eet
woe i
Hil Direct sound
Diffuse sound
C. Distribution of direct and diffuse sound in an enclosed space
Application Concepts Secondary concepts Short definition

area
Speech and listenability . speech listenability acoustic suitability of a room

music —.
fo music listenability for speech (1) or music
performance (2)
Music transparency . fange transparency clarity of a musical

— .
|Pw time transparency performance through ability to
differentiate simultaneous (1)
or sequential (2) sound
events.
Speech and room = . listener involve- sensation of the size and

music involvement ment treatment of a room
. room size
. echoiness
. roominess
OOM
D. Concepts of aural acoustics

Room acoustics, fundamentals 9
The listenability of a room generally describes its suitability for certain sound
events. Good listenability of a room for speech, for example, means that one is able
to hear speech well at every seat in the house without the need for sound reenforce-
ment. The judgement of listenability is obtained psycho-acoustically and combines
different aspects of aural acoustics. It is dependent both on the nature of the sound
source and the very subjective criteria of the person who is judging.
The transparency for musical offerings denotes Jone’s ability to differentiate
between simultaneously played aeay se ae pana groups or their various regist-
ers, in spite of superimposed room reverberation. Transparency is a basic requirement
when one wishes to hear complex musical structures. (One also terms as transparent
the clear separation of musical elements which follow one another in time) (time
transparency). Transparency, therefore, refers to the clarity of a musical offering; it
is comparable to speech intelligibility, meaning one’s ability to comprehend every word.
usical sound reflections which arrive no later than 80 ms after the beginning of a
sound event increase transparency and the sensation of ee arriving
later adversely affect transparency and increase reverberation. For speech, this time
boundary is best placed at 50 ms. i
The room impression is the aural sensation one obtains in a partially enclosed
(courtyard) or completely enclosed space after the sound has commenced. The room
impression has several components: (1) The sensation of being in the same room with
the sound source, and not, as is the case with stereo reproduction, of listening into the
recording room as through a window; (2) The sensation of the room size, especially
its width and depth; (3) The sensation of echoiness, meaning that diffuse sound is
present alongside the direct sound without seeming to be a repetition of the direct
sound; (4) The sensation of spacial impression, meaning the sensation that sound
emanates from a larger room area than would correspond to the size of the sound
source. It is caused by early reflections (10-80 ms) and may be perceived only during
reproduction at higher sound levels (75-85 dB). By contrast to echoiness, an echo is
perceived as a repetition of a sound event and provides the listener with a clue about
the distance and nature of a wall located at a greater distance from the observer.
[Significance of room tone for microphone placement and the listening expeuenee|
The microphone generally picks up both direct and diffuse sound. While the
direct sound is influenced little by the nature of the room, the diffuse room tone
transmits information about the room size and the nature of the wall treatment. The
room integrates the sound which is radiated into the various directions, as for instance
from a musical instrument. It therefore does not represent a specific sound aspect of
the instrument, but rather something which may be described as total sound. When
we listen live in a room, we predominantly get the room tone, and it therefore sounds
natural. The acoustical attributes of the room tone provide information about the
cultural and social environment into which a musical performance has been placed.
For music it represents something similar to the room into which a sculpture has been
placed and its interior design. Thus church music requires the acoustics of a large
church for which it generally is written; symphonic music is written for concert halls,
chamber music for the small, private room in a castle or home. Folk music needs the
intimate atmosphere of a peasant inn or the out-of-doors. In the area of pop music
and other similar musical forms, we see the creation, through the use of artificial
reverberation of new acoustical surroundings which really do not exist in real life.
Even for pop concerts, whether indoors or outdoors, the existing room acoustics are
eliminated through the use of intense sound systems, and are replaced by new, non-na-
tural acoustical surroundings. ™ == “~—™
Direct sound
The sound wave or direct sound experiences several changes on its direct path
from the sound source to the microphone or listener. The so-called geometric propa-
gation attenuation leads to a level reduction with increasing distance from the sound
source, depending on its size and nature. Additional attenuation is provided by the
air, depending on its humidity and temperature, on obstacles and on sound absorption,
as for example propagation over an audience or empty rows of seats.
Geometric propagation attenuation

For point sound sources, i.e., all sound sources which are small when compared
to their radiated sound wave length, the sound energy is distributed on spherical shells
whose surface areas increase rapidly with increasing distance; they grow as the square
of the distance. Because of the fact that the sound energy has to be distributed over
an ever larger surface, the sound pressure or level also is greatly reduced with increas-
ing distance. With every doubling of the distance, the sound pressure is cut in half,
i.é., the sound pressure level is reduced by 6 dB (fig.A).} This reduction therefore is
most pronounced in the immediate vicinity of the sound source and insignificant at
great distances from the sound source. For a line sound source, as represenied by
traffic on a highway or a railroad, the sound radiates on cylindrical shells; the sound
level attenuation is less and therefore amounts to only 3 dB for every doubling of the
sound source distance. It is for this reason that streets and railroads may be heard
at remarkably great distances. {For large area sound sources the level initially is
hardly reduced at all7] Beginning at a certain distance, which is a function of the
dimension of the sound radiating surface, every surface of finite dimensions becomes
a non-directional sound source, and the sound level then attenuates more rapidly.
Bellis, such as are found on brass instruments, behave similarly. The propagation
attenuation of real sound sources such as musical instruments is dependent on the
particular instrument’s directional characteristics (+p.31). Experience shows that the
average attenuation, up to a distance of several meters, equals 4 dB per distance
doubling. In view of the fact that real sound sources may best be compared to surface
radiators of limited dimensions, there is a frequency dependent radiation attenuation.
In general musical instruments may be assumed to be surface radiators at higher
frequencies while they may be considered point source radiator at low frequencies.
This results in a more rapid attenuation of the low frequency components than of the
high frequency ones with increasing distance from the sound source. This effect is
reenforced both for the listener and the microphone because of the lower sensitivity
of the ear for low frequencies with reducing sound level.
In enclosed, not anechoic, rooms the direct and diffuse (reverberant) sounds
intermingle. Only within the effective reverberation radius (+p.26), whose magnitude
is a function of the room acoustics and the directionality of the sound source and
microphone, does the direct sound portion predominate.
Influences of temperature, humidity, and wind

The climatic influences on sound propagation are relatively complex and affect
only great distances and the reverberation, where even for medium reverberation
times, the sound waves traverse distances of several hundred meters (+p.22). Generally
speaking the air propagation attenuation increases with frequency and is added to the
geometric propagation attenuation. ([n the center of the response range, the attenua-
tion decreases with increasing relative humidity; likewise with rising temperature.
Direct sound if
A. Geometric propa- r ar 4r br
gation attenuation 0 :
distance from
sound source
1. point sound source (r = reference

2. line sound source distance)
3. plane sound source
(curve is function 6
of plane size and
distance)
¢
12
dBy rel. sound level
r 2r Si 4r or 6r TAF 8r
0 distance from
sound source
(r = reference
distance)
WZ
18
dB ¥ rel. sound level
B. Influence of - wind
wind on sound ==————
propagation = ae
=
acoustic horizon sound rays

C. Diffracting effect level

of an obstacle
and its influence
on the frequency
response
S Hl |
frequency
ae
mu oI | || |
level
frequency
D. Estimation of the size of the obstacle

diffracting effect 30 teens
of obstacles 20 Nivea
m seem -
10 ane
5 Seine are shadow area
fete eee
ors
“= frequency
30 50 100 300 500 1000 3000 10000 15000 Hz
E. Direct sound fre-

quency response attenuation compared to unimpeded propagation
at a distance of
about 20 m (66 ft)
after passing over yp
rows of seats
20
100 500 1000 Hz frequency

Direct sound 13
These changes, however, are relatively minor. Considering today’s air conditioned
studios and concert halls, these influences are no longer of any practical significance.
Out-of-doors these influences are joined by the effect of wind. In view of the fact that
wind velocity increases with increasing altitude above the terrain, sound does not
radiate in a straight line. [Against the wind direction, the sound is deflected peaeecs
with the wind direction, downward towards the earth\fig.B). The result is that soun
is reenforced when traveling inthe direction of the wind, and greatly attenuated against
the wind with an acoustic horizon which appears to rise with increasing distance from
the sound source. Below the acoustic horizon, the sound may well be attenuated by
as much as 30 dB.
Sound diffraction around obstacles

~ Ofttimes there are obstacles such as columns or people between the sound
source and the listener or microphone. The resulting sound shadow, by contrast to
those cast by a light, are not sharply outlined, but rather are areas with greater or
lesser soynd coloration. Sound partials whose wave length is small when compared
to the dimensions of the obstacle, are diffracted. If the wave length is small when
compared to the dimensions of the obstacle, a shadow is created behind the obstacle.
The transitions are gradual. Therefore, the result of an obstacle in the sound field
is a dulling of the sound, and that increasingly so the larger the obstacle (fig.C). Fig.D
permits an estimation of the shadow effect. It is usual to use separating screens
(gobos) during recording as obstacles for separating sound waves (+p. 94).
So absorption for sound radiation over an audience seating area

If sound travels over an absorbing surface such as an audience, sound energy is
extracted and results in an additional attenuation of the sound wave.) Since absorbers
usually are effective only over_a small portion of the frequency spéctrum, the sound
assumes a certain coloration. isurtiess with plantings display increasing attenuation
towards high frequencies {Sound propagation over an audience area or over empty
seating areas shows a severe dip in the area of 100-300 Hz, which may be as much as
10-20 dB at a distance of 20 m (66 ft) from the stage (fig.E). This is true for the
height of the heads, above which the effect abates rapidly. [A sufficient amount of
energy supplied from ceiling reflection to the rear seat rows can largely compensate
for this effect. *)
The meaning of direct sound in recording

Generally speaking, the microphone receives) both direct sound and rogm tone
(diffuse sound). In most cases the direct sound predominates. The following sound
properties may be determined only from an analysis of the direct sound:/the rise time
behavior and the random, noise-like sound components are present in their true
nature only in the direct sound.) They are largely responsible for the sound presence.
The timbre of directionally radiated sound changes in the area of predominantly
direct sound with the location of the microphone. It is only here that an influence of
the sound timbre through microphone location is possible. The information about the
localization of the sound source in the room, in other words its localizability with
regard to direction and distance, is possible only when a sufficient amount of direct
sound reaches the listener or microphone. On the other hand, the direct sound,
especially for musical instruments, always represents a very specific sound aspect, one
which exists in this form only in a specific direction and at a specific distance. For
this reason direct sound provides the possibility of sound manipulation, by replacing
one sound aspect with another.
Sound wave reflections
Creation of sound reflections

[Sound reflections within a room occur when sound reaches a boundary surface
without too much absorption:? Dimensions and nature of the surface determine
whether the reflected sound i§ to be reflected in only gne direction, similar to light
from a mirror, or isto be diffused or scattered. For the former to occur, the surface’s
dimensions must be at least several wave lengths (fig-A). In view of the fact that
wave length increases with decreasing frequency, it would require a surface of ve
large size to reflect the lowest frequency, while smaller reflecting surfaces reflect only
higher frequency components, so that the reflection sounds brighter and tighter than
does the direct sound. {Directed sound reflections are used in room acoustics in the
stage area for better communications between musicians and in ceiling design for the
purpose of providing adequate acoustical energy to areas mast distant from the stage.
The laws of reflection are analogous to those laws found in the reflection of
light: angle of incidence equals angle of reflection. A corner reflects the sound back
to the direction from which it emanated due to double reflection, but with a certain
time delay (fig.B). Curved surfaces behave either as concentrating or diffusing mirrors
(fig.C). In practice arched ceilings lead to unexpectedly strong sound concentrations,
which can lead to serigus sound falsification. But even flat surfaces may cause
disturbing reflections; for example the rear walls of auditoriums. We call first reflec-
tions those reflections which arrive immediately following the direct sound. They
have a very significant roles to play in the acoustic behavior of a room (-p.6).
Effect of sound reflections

Even though we perceive sound reflections only as such if they arrive after more
than about 50 ms, they have a very significant effect as first reflections—as such not
consciously perceivable—on the sound picture. [These first reflections together with
reverberation, provide a significant sound clue to the room’s interior, its size, its
proportions and wall treatment,)and out-of-doors, about the listener’s surroundings.
This acoustic room sensation is superimposed on the music, speech or noises. The
most significant attributes of single reflections are (fig.D):
True and false sound source localization

ven though the directions from which reflections arrive normally do not cor-
respond to the direction of the direct sound, the ear localizes the sound source using
the direct sound because, for virtually identical sound events, the ear analyzes always
the wave front arriving first as to its source direction. The precedence effect, also
called the Haas Effect, explains this phenomenon./Only if the direct sound is atten-
uated to less than half loudness through possiblé obstacles in its path (level of the
direct sound more than 10 dB below the level of the reflection) is there a false
localization of the sound source. In such a case it is the reflection which determines
the direction from which thé sound appears to come. This effect often applies only
to single instruments, often even to single notes of opera orchestra instruments for
listeners in an opera house.
Increase in sound source loudness level

Reflections with a delay of less than 15 ms can raise the apparent sound level
by several dB. The sound clarity decreases with increasing reflection delay,(especially
in the presence of reverberation. ) i ve = 7 =
Sound reflections 15
extent of the
reflecting surface
no sound
reflection
A. Necessary size of
reflecting surfaces
as a function of
frequency
30 50 100 300 1000 3000 5000 15000 Hz
frequency
angle of _ angle of
incidence reflection
r/2 le
we —o oO
ee
The sound is dis-

—— persed if the sound
direction of incidence = — source is close to
the sound mirror.
direction of reflection
The sound is re-
flected back ex-
actly to the sound
source if it is lo-
cated at the sound
mirror radius.
The sound is reflec-

ted in parallel rays
if the sound source
B. Sound reflection from is located at a dist-
plane surfaces and ance equal to half
corners the sound mirror
radius.
If the sound arrives

in parallel rays, it
will be concentrated
C. Sound reflections at a point equal to
from curved half the sound mirror
surfaces radius.
D. The effect of mis-localization _ | yh j is

reflections for strong reflections Effect
| especially
| for direct
incr. listening
loudns
sound
coloratn
Effect
| especially
| during
recording
medium
room
— Effect
for arallel for stron especially
walls single re ections + when list-
sound flutter echo ening and
— recording
reflection
Om no Rees 10 14 20124527. S034. ime detour
oo
0 10 20 30 40 50 60 70 80 90100 ms reflection
delay
level
E. Frequency response dB
of the super impo- +6
sition of a sound
wave on its own 0)
reflection; here frequency
—6
at equal level
(comb filter curve) 12
=e
—24
F. Standing waves
between parallel
walls
(A= wavelength).
Sound wave reflections 17
This loudness level increase is only of significance to a live listener, and is of no

significance to a recording. On the contrary, such reflections may impart unfavorable
sound coloration.
Sound coloration
Reflections with a delay of between 1 and 15 ms—usually floor reflections —lead
to disagreeable sound coloration in recordings. These result from cancellations and
augmentations within the response range, {resulting from the superimposition of the
direct sound on its reflections. }The result is a frequency response corresponding to
that of a comb filter curve, which, through periodically alternating maxima and minima
with several dB level difference, show similarities to the construction of an harmonic
sound (fig.E). The result is a type of coloration with a pitch characteristic. For a
strong reflection in a musical instrument we get a barrely sound. This coloration
becomes extremely disagreeable if the sound source is in motion. In such a case the
pitch characteristic changes with a sound something like wiuiui and is akin to the
sound caused by multi-path reception of short wave broadcasts. This effect is often
generated electronically. It is then called phasing and is used as an effect in the
production of popular music recordings. This coloration is particularly disturbing to
speech, especially when the speaker is reading at a table (+p.126), as well as when
using support microphones in musical recordings (+p.104).
Apparent room size and spacial impression

(The subjective impression)one obtains in a room is composed of the sensation
of room size, the echoiness and the spacial impression of the sound source. The
impression of rogm size is predominantly imparted through individual reflections
(early reflections). Their delay with respect to the direct sound is a measure of the
distance of the reflecting wall and imparts a subjective sense of the room size. For
a delay of under 10 ms, corresponding to a round trip sound path of 3 m (9.8 ft), an
impression of a small room (living room) is obtained, besides the coloration obtained
in a recording. This is especially true when the direction of the direct sound and its
reflection overlap. [The room size impression increases with increased reflection
delay. ?A small room corresponds to a delay of between 10-25 ms; a medium size
room a delay of between 25-50 ms, and a large room a delay of between 50-100 ms.
The distinct sensation of room size increases with increasing level, until a limit is
reached as a function of the delay time, above which the reflection becomes disturbing.
For music recording this boundary value is higher than for speech recording.
Echo, flutter echo, tailing echo

Individual reflections with a delay of at least 50 ms, corresponding to a round
trip path length of 17 m (56 ft) and of sufficient level, appear as an echo. An impulse
sound between two parallel, sound reflecting surfaces is bounced back and forth,
generating a so-called flufter echo. For a wall spacing of a few meters, these sound
impulses bounce back and forth so rapidly that this sequence of impulses acquires a
pitch characteristic which produces a tailing echo which attaches itself like reverbera-
tion to every sound impulse. Continuous sound, and this includes even individual
tones in music, often produces very disturbing standing waves between two parallel
walls (fig.F). They often amplify individual bass notes inordinately (above all organ
tones) if the distance between the two walls is half the wave length or a multiple
thereof. Since the pressure maxima and minima of a standing wave remain stationary,
{a minor displacement of the microphone will usually bring relie
tandig waves significantly determine the acoustics of small rooms if the walls
are not highly absorbent. (The result is erratic behavior in the reproduction of the low
frequencies. }
Absorption
[teacoustical properties of a room are influenced, aside from its geometric form,
by thé sound absorbing characteristics of its boundary surfaces.) Together with the
room volume it determines Both the reverberation time and its coloration. The
audience forms a significant sound absorber. When sound energy in the form of a
sound wave impinges on a sound absorber, part of the sound energy is extracted from
the sound field by the absorber, while the rest is reflected back into the room. The
absorption coefficient indicates what part of the impinging energy is absorbed by the
wall; a coefficient of 1 means that no sound is reflected and that the material is soft.
A coefficient of 0 means totai reflection; the material is hard. Every partially absorb-
ent surface maybe equated to a surface with total absorption. An open window is
a total absorber. As a result, every absorbing surface may be expressed as so many
m? open window. A part of the energy which is not reflected generally passes through
the wall into the adjoining room. The other portion is extracted from the sound field
through conversion to heat. The total absorptiveness of a room, its so-called absorb-
ability, is made up of the sizes of the individual absorbing surfaces and their individual
absorption coefficients, or from the equivalent area open window. Sound attenuation
or the sound absorption of a wall interferes with sound propagation through withdrawal
of sound energy while sound reduction does the same thing through reflection from
obstacles. The effectiveness of absorbers is highly frequency dependent. There is no
such thing as a universal absorber which is equally effective for all parts of the
frequency spectrum. This is a result of the great differences between the wave lengths
of low frequency tones up to about 20 m (66 ft) and high frequency tones about 2 cm
(0.8"). The acoustical mechanisms which lead to sound absorption operate accord-
ing to vastly differing principles depending on the different wave lengths.
High frequency absorbers

~ High frequency absorbers are of eeuseedlaltee cies upwards of 500 to 1000
Hz, depending on the material of whic ey are madé (figA). Purely high frequen-
cy absorbers color the reverberation dark making the room sound dull and tubby.
For this reason they are mostly(used in modified form prGn combination with low
frequency absorbers.) High.frequency absorbers are principally madeof porous
matgyial. The absorption results from the fact that the velocity of the air particles
is damped by the porous material. For this reason only open pored materials, such
as fibrous ones, are suitable. The moving air particles cannot enter plastic foams with
closed pores, and therefore such materials are absorptive only over a narrow band of
frequencies. Generally speaking it is impossible to estimate the absorption properties
of such materials. e absorption properties of porous materials increase with
increasing thickness.) For one thing the absorption coefficient rises and for another
its effectiveness expands towards lower frequencies. There is an optimym thickness
of material for every frequency and any increase beyond this thickness will not improve
the absorption. An air space between the absorber and the wall behind it extends the
absorptiveness towards low frequencies, something which is always desirable. Examples
of porous absorbers are soft mats made of textile, glass or rock wool (only at relatively
high frequencies) and spray applied atop porous layers such as asbestos (no longer
used.) A layer of paint over such a porous absorber may close the pores and thus
adversely influence its effectiveness. Curtains, especially at a distance from the wall
Absorption 19
3
porous layer curtain with folds typical reverberation
frequency response
reverberation time
frequency
100 1000 5000 Hz 100 1000 5000 Hz frequency

High frequency absorbers
porous layer with typical reverberation

perforated overlay acoustic panel frequency response
absorption coefficient
reverberation time
porous layer without
solid overlay
frequency
100 1000 5000 Hz 100 1000 5000 Hz frequency

Mid range absorbers
vibrating panels resonators typical reverberation
ef
IUD \CLI frequency response
74, mSrag,
eas oa
absorption coefficient reverberation time
‘Gas
frequency
100 1000 5000 Hz 100 1000 5000 Hz frequency

Low frequency absorbers
A. High, mid and low frequency absorbers and their effect on the frequency
ad
response of the reverberation time
eee
]
Absorption coefficient
0,8
100 200 500 1000 2000 Hz 5000

frequency ——=—
audience on wooden seats
cushioned seats
audience on cushioned seats
Sih
B. Absorption coefficient of seating and audience
Absorption coefficient
100 200 500 1000 2000 Hz 5000
1. smooth plaster on wall frequency ——e

2. suspended, plastered ceiling 5. parquet floor
3. 16mm (0.7") thick wood with 6. carpeted floor; medium thickness
porous backing filler 7. curtains; medium thickness
4. 7mm (0.3") thick plywood with
porous backing filler
C. Absorption coefficient of wall and ceiling materials

Absorption Di
and hung in folds, as well as carpeting and rugs may be very effective (fig.C, also
noting additional examples). Air is also a high frequency absorber, but with only
minor effectiveness (+p.10). = 4 sF
~
Further important high frequency absqrbers are seating and an audience (fig.B).
Seats have been designed which have an absorption coefficient similar to that of a
person sitting in it, in an effort to minimize the difference between rooms with and
without an audience. As a result, the room acoustics of modern concert halls and
studios do not change significantly with and without an audience. In chyrches we find
this difference to be relatively large, with increasing significance when the room
volume per person decreases. The same obviously holds true for other rooms which
are not equipped with padded seating.
Low frequency absorbers

ow frequency absorbers are effective in the frequency range below 300-500 H
while their effect on high frequencies is negligible. There exist two types of absorbers:
Helmholtz resonators, and resonant panels. (Enclosed air spaces which are coupled
to the sound field through small openings, act as Helmholtz resonators (fig.A). The
second type of low frequency absorber, the resonant panel, is architecturally particu-
larly advantageous for it permits the use of wall paneling as low frequency absorbers.
These panels are free to vibrate in front of an air cushion, e.g., an enclosed air space,
which may be filled with loosely packed porous material for increased absorption. The
use of thicker plates, or the application of weights to the panels, or an increase in the
volume of the air cushion behind them, permits their effectiveness to be expanded
towards lower frequencies (fig.C: 2,3,4). Small resonant absorbers along the edges of
a room multiply their effectiveness by two, and in the corners by four. The combi-
nation of low frequency absorbers (for example wooden wall panels) and high frequen-
cy absorbers (for example carpeting, upholstered chairs or curtains) generally provides
favorable reverberation characteristics together with a pleasant increase in rever-
beration time at mid frequencies, while at the same time providing a pleasing interior
decorating solution.
Mid range absgrbers

Mid range absorbers, especially ones effective between 300 and 1000 Hz, are
created through the combination ofporous absorbers) and(resonating panels of wood
or plaster. ) The panels are slotted or perforated to permit the porous layer behind
them to become effective. Acoustical panels are manufactured absorber components
of various constructions; usually pressed fiber panels or boxes with perforated or
slotted front covers which generally act as effective absorbers in the mid and high
frequency ranges. They are used both to reduce the reverberation time and to reduce
room noise. ie
Minimally reflecting rooms

These rooms (often incorrectly called reflection-free or anechoic) exhibit almost
solely a direct or free sound field, due to the fact that their walls are covered with
highly effective wedge-shaped absorbers made of rock wool or glass fibers. The
wedge shape provides a gradual transition between the air and wall. The absorption
coefficient for higher frequencies is 1; the absorption effectiveness is total down to
about 100 Hz. Minimally reflecting rooms have a reverberation time of about 0.1 s.
They serve as acoustical test chambers, for example for obtaining the directional
characteristics of a Microphone or musical instrument. ee mr
In an effort to provide the environment of an out-of-doors scene in radio drama,
film or TV sound, it is common to use specially constructed studios which are often
described as anechoic, but which nevertheless have a reverberation time of under 0.2 s.
Such studios have their walls completely covered with thick layers of fibrous panels.
22. The recording environment
Reverberation
The phenomenon reverberation

Reverberation adds a type of after sound to every sound event and this greatly
influences the characteristicof the sound. Pauses in music are, as a result, largely
filled in (fig.A) and replaced with information about the particular recording space.
Reverberation is composed of a great number of reflections which become weaker
and increase in their frequency the later they arrive at the listener or microphone
(+p.6). The reverberation of a sound event reaches the listener or the microphone
after a certain amount of delay. In a room in which the total absorption is almost
evenly divided between walls, ceiling, and floor, the acoustical structure of the rever-
beration is virtually identical everywhere in the room. Those reflections which arrive
earliest have a special significance for the recognizability of the room and its structure;
especially the room size (+p.14). [Reverberation must be considered as having three’
components: everberation build-up,quasi stationery state,)and(decay Yfig.B). If the
build-up maximum is reached in under 50 ms, speech sounds distinct but music sounds
rather hard. Musicians on stage need such a relatively short reverberation build-up
in order to hear one another and thereby to play together. A long reverberation
build-up makes speech indistinct and the attacks in music rather soft. |The quasi sta-
tionery reverberation, increases the loudness of a sound source in a hall. | It also
provides for an intermingling of successive tones in music, a function which i§ analog-
ous to the function of the sustaining pedal of a piano. Another important function
is the apparent enlargement of a sound source as a result of early reflections, some-
thing which is perceived as an agreeable spacial impression of a performance (+p.6)
provided that strong reflections from the direction of the sound source do not destroy
this impression. Frirst reflections and reverberation decay inform the listener about
“aagsipenandalnictian treatment of a room.] It involves the listener in the sound event
and its spacial development.
Properties of reverberation
Reverberation time and frequency response are its most important properties.
(Reverberation time)is defined as the time which elapses after the sound source has
ceased, until the sound level has dropped by 60 dB. This corresponds to a drop in
sound energy to one one-millionth of its initial value. Reverberation time is measured
by extrapolation from a level drop between -5 and -35 dB below maximum level
(fig.C). A true level drop of 60 dB can only be perceived in practice in pauses after
a loud orchestral passage. For this reason,the subjectively perceived reverberation
duration is a function of the loudness of the sound source.The perceived reverbera-
tion duration only comes within range of the measured reverberation time for high
loudness levels and in quiet halls. The lower the loudness level, the shorter the
perceived reverberation duration. The reverberation time is normally given for the
empty hall, the occupied hall, or for the hall with only the orchestra on stage. If only
one value is given, it refers to the frequency range between 500 and 1000 Hz. Normal-
ly, however, the reverberation time between 100 and 8000 Hz is given by means of a
response graph. [the reverberation time is a function of the room volume and the total
absorption:
-_
e S
Reverberation time (ins) = 0.163 - room volume (in m’)
total absorption (in m* open window)
Reverberation 23
sound level
i time
ae
=
A. Effect of reverberation on the sound level for the
performance of music
radiated sound
source power
rel. sound level
energy density
in the room
test time
| ! time reverberation time time

J
build-up steady decay
state
B. Build-up, steady state and decay C. Measurement of the reverberation time
—. wh Ss)
reverberation time
5,0
So
aoeSES
4,0
3,0 SesSe i SSI
SKS OD sees VUS
20 Yl eee SESS oe
d dd
l
LMdlddddddtea bb zm
= m
mm TT
a8 N GGG ———
AQ
'N’A’AYSSSA SSS SSSSS
0.6RSSASASHHSSSS
OOM:
0,3
room volume
100 m° 1000 m? 10000 m° 100000 m?
D. Favorable reverberation times for various room usages
A.Pop music studio D. Concert hall with audience

B. Conference room E. Church, empty (max.)
C. Opera house with audience
reverberation time
reverberation __ reverberation
dull mid-range emphasized
reverberation rela-
tively neutral sounding
E. Typical reverberation 100 1000 10000. Hz

frequency response frequency
— el
Reverberation D5
Room volume increases with increasing room size more rapidly than do the
room’s absorbing surfaces. For this reason(reverberation time increases with room
sizejeven for identical wall treatment, although only with the room volume’s cube root.
As an approximation: the room reverberation time increases proportionately to its edge
length. The reverberation time will double for a hall with eight times the volume!
e longer reverberation time of large rogms amplifies the sound source loudness
which is indeed a requirement for large halls. It is impossibleto make any clear
comment about the ideal reverberation time, because subjective judgments and
objective requirements find little or no consensus. However, it is possible to state the
upper and lower boundaries of preferred reverberation times (fig.D). Desirable
reverberation times, on the one hand, are a function of room volume, since larger
rooms always require longer reverberation times. On the other hand, the room’s
purpose requires specific reverberation times. Short, but not too short, reverberation
times are needed for speech to provide the required intelligibility, while music requires
longer ones, depending on the style of music in question. Old churches have the
longest reverberation times due to structural properties. In more moder churches
the reverberation time was kept shorter for purposes of greater intelligibility. Multi--
purpose rooms, therefore, must be problematical.
Aside from the average reverberation time,(the frequency dependent nature of
the reverberation time is its most important attribute.) Frequency dependent behavior
stems from the fact thatthe absorption of walls, ceiling and floor are nearly always
frequency dependent. LAdded to this is the high frequency attenuation caused by air
absorption. \The result is that the reverberation has a quality quite distinct from that
of the direct sound. Due to the current practice of air conditioning concert halls and
studios, temperature and humidity have only a minor influence on reverberation.
Rising temperature and humidity may slightly change reverberation characteristics.
Because low frequency absorbers are costly, and additionally an audience, carpeting,
curtains etc. absorb predominantly mid and higher frequencies, rooms which have not
been acoustically planned, especially smaller ones, display an increase in low frequency
reverberation time.} A relatively frequency independent reverberation time, or if
anything a slight rise in the mid range, has been found to be advantageous for record-
ing studios. Good concert halls may display quite different frequency dependent
reverberation times, since a good concert hall is not solely defined by its reverbera-
tion characteristic. It is interesting to note that many halls which are considered
excellent display a mid range reverberation time accentuation.
Often artificial reverberation must be added in the recording process, because
there is insufficient natural reverberation, or because of a desire to provide increased
presence by placing microphones in close proximity to sound sources. This largely
eliminates natural reverberation and brings into play the manifold possibilities which
are provided by artificial reverberation. Such equipment may consist of plates, springs,
foils, and electronic units. Modern electronic reverberation equipment operating in
the digital domain provides the adjustability of reverberation time and frequency
response, the addition of early reflections, as well as delaying the reverberation onset.
Many reverb effects such as extra long reverberation times and unusual frequency
response characteristics exist only synthetically. Often such units permit special
acoustical simulation settings. In popular music recording nowadays there only exists
the artificial room. For serious music, where excellent room acoustics are a prere-
quisite, artificial reverberation can only augment natural reverberation to a limited
extent. The reason for this is that reverberation time and intensity largely help to
determine the interpretation, especially the tempo and the articulation, meaning the
separation or joining of the individual tones. (Reverberation should influence the
clarity of music as little as possible.) Artificial reverberation always reproduces the
special sound aspects in a specific sound radiating direction, while ee reverberation
reproduces the total sound radiated in all directions.
Reverberation radius
- Relationship between direct and diffyse sound

A sound source within an enclosed space builds up a direct sound field around
itself whose sound level attenuates rather rapidly with increasing distance (+p.10).
It furthermore creates a diffuse sound field (reverberation) which, in the ideal case,
is the same everywhere in the room (~p.22). {Therefore at every position within the
space there exists a specific ratio between direct and diffuse sound, the so-called
reverberation ratio, or more properly, the direct sound-to-reverberation level ratio.)
Since itisa requirement in recording that the direct sound portion of the sound wave
should predominate, and for proper localization in stereo recording must predominate,
it is necessary to have complete knowledge about that area around the sound source,
in which the direct sound is greater than the diffuse sound. At the distance from the
«# sound source which we call the reverberation radius, the direct and diffuse sound
levels are equal (fig.A). Generally speaking, microphones should be set up within the
reverberation radivs. The reverberation radius increases with the square root of the
room~~» volume, and(decreases)with the inverse square root of the reverberation
$e
time:
t
Gis TE
: nae room volume (in m
Reverberation radius (in meters) = 0.057 pogmivo eanEA eee
reverberation time (in s)
Since the reverberation time, with its usual values between 0.5 and 2.5 seconds,
-~¥ varies relativel little. (itis especially the room volume which determines the rever-
beration radius \(fig.B). In other words, the larger the room, the greater the distance
atwhich the microphone may_be positioned from_the sound source. For this reason,
one usually selects a larger room for large musical groups rather than a small room
of equal reverberation time, since it permits greater sound source to microphone
distance without the need to give up any presence or localization. The values shown
in fig.B show that the reverberation radius is surprisingly small. In practice, these
values must undergo significant corrections (see below), since they apply only to
omni-directional sound sources and omni-directional microphones. Only non-direct-
ed, spherically radiating sound sources have their reverberation radii, in all directions,
lie on a spherical shell. Fig.C shows the typical relationship betweendirect and diffuse
sound for small and large, dry and reverberant rooms for a given microphone distance.
Reverberation radius corrections in practice

Under practical conditions, the reverberation radius is usually considerably larger
than the formula indicates. Reason for this is {the usually directional nature of sound
sources, Jand [the preferentially used directional microphones} In practice it is often
difficult to estimate the reverberation radius. One simple method consists of placing
a microphone relatively far from the sound source, and to move a second microphone
from this distance ever closer to the sound source until the level difference between
these two microphones reaches 3 dB. The reverberation radius is at this second point.
The microphone type selected for the recording at hand is to be used for this test.
Reverberation radius correction due to the microphones’ directional characteristics

In practice, the reverberation radius is larger than given in fig.B. In directional
microphones, which partially discriminate against sounds from other directions, the
reverberation radius may attain values of up to twice the calculated value, depending
on the directional characteristic and its frequency dependence (fig.D).
Reverberation radius NY)
A. Definition of rel. sound level

reverberation
radius
direct sound level

= diffuse sound level
direct + diffuse sound

(sound field within room)
direct sound
Vy distance from sound source

reverberation radius
B. Reverberation
radius for various reverberation time
room sizes and
reverb times, valid ms HO er kt 20 Bey 25 80 B65 40s
for omni micro-
phone and omni- 500 |1.27 1.04 0.96 0.9 0.85 0.8 0.74 0.68 0.64
directional
sound source ®o 1000) 1:85 -1.47 1.35 141.2 1:14 1.04 096 0:9
fe
3 PAGO Eo els) alley abs} alae IS ez eS eae
>
E 5000)|4,03' 3/29 3.04 2.85 2.691255 2338 216, 2:02
£40000 AG5AB: 403-3. 8S Ghes Se 6.058255

15000 5.28: 4,98. 4165 4427 403° 3:78 3,49
20000 Oh SSH isl A465 4.3) 4,08
C. Typical level re- small room large room

lationships of aa - aN
direct and dif-
r # sound level sound level
fuse sound for
identical micro- short
phone distance rever-,<
beration
and for large
time
and small rooms
and short and
long reverbera- sound level time
tion time long
rever-
beration
time
Peat
fH
ra
(1) direct sound

E=} diffuse sound
figure-8
. \
\
\hyper-cardioid
CG Aicbees
Or
cardioid
\ ph
| ; | cardioid
| cS uper-cardioid
1, 0 Go Hew I) ZO) 25 2x2
/ / Ll |
increase of
sound source
(omni-directional) reverb radius
i}
by a factor
D. Increase of reverberation radius as a function of microphone

directional characteristic
Increase of reverb radius

by a factor
tuba trombone
Clarinet
trumpet
oboe
frequency
0,1 0,2 0:3 0,5 1,0 2,0 3,0 5,0 kHz 10,0
E. Increase of reverberation radius as a function of musical instrument directional

characteristic in its principal radiation direction
Reverberation radius P48)
The factor by which the reverberation radius is increased is found by applying the
square root of the directivity. The directivity, and with it the increase in reverberation
radius, is typically frequency dependent: cardioids, super-cardioids, and figure-8
patterns are largely frequency independent. On the other hand, pressure transducers
(omni microphones) transition into cardioid and uni-directional patterns towards the
upper end of the response range, which means that the reverberation radius doubles
for these microphones above 8-10 kHz._In actuality this generally providesa desirable
presence effect for good recordings. [The only type of microphone which in no way
increases the reverberation radius, i.e., the type which accepts the greatest amount of
diffuse sound at its location, is the boundary surface microphong (+p.74). For stereo
coincidence microphones one can obtain the enlargement of the reverberation radius
from the directional characteristic of the mono (sum) signal (+p.90) in combination
with the data found in fig.D.
Reverberation radius correction on the basis of the directional characteristic of the sound
source
By contrast to the effect which the directional characteristic of microphones has
on the reverberation radius, the directional behavior of sound sources, especially those
of musical instruments, is so complex, that it is impossible to describe it using a simple
numerical value. (A measure for the change of reverberation radius at a specific
frequency in a specific direction is the so-called statistical directivity factor.) It
indicates by what factor the reverberation radius from ig.B must be multiplied to
obtain the actual specific sound source reverberation radius for a specific tone and in
a specified direction. The directivity factor is not only a function of the type of
instrument, but obviously depends very much on the frequency and the emanating
direction. Generally it is a fact that every instrument may be viewed at low frequen-
cies as an omni-directional radiator, i.e., does not have an enlarged reverberation
radius. (The reverberation radius increases with increasing frequency and in the
direction of maximum radiation: In other directions, however, it will be reduced
significantly from that of an omni-directional radiator. Brass instruments have extreme-
ly high directivity factors at higher frequencies in the direction of their bell (fig.E),
because of the fact that the bell produces a sharply directed sound signal. The
directivity factor may vary greatly even for small changes in radiating angle.
The actual increase in the reverberation radius is the product of the factors
attributable to both microphone and sound source. It is therefore possible that, in
practice, and especially for the principal direction of brass instruments, the entire
room stays within the’ reverberation radius. In practice, the frequency dependent
characteristic of the reverberation radius increases the reverberation radius at higher
frequencies, and thus improves the presence effect in a desirable manner. Directivity
of microphones rapidly decreases beyond the corrected reverberation radius. There
is no localization possible in the diffuse sound field.
Audible sensation and reverberation radius

The human ear is well equipped to distinguish changes in the ratio of direct to
diffuse sound. Changes in the reverberation level of as little as 2 dB are clearly per-
ceived. The ear is generally very sensitive to room acoustics and its changes. Any
change in distance from the sound source in a room with considerable reverberation
is determined from the ratio between direct and diffuse sound. Because of this great
sensitivity, it is possible to perceive distance changes of only a few percent, and
therefore distance judgement is much more acute in enclosed spaces than outdoors.
Given a sufficient amount of direct sound, the ear senses the location and spacial
extent of the sound source with the arrival of the first wave front. The early re-
flections and the reverberation provide to the ear information about the space, while
at the same time the sound source dimension grows until it fills the entire space.
ae ae fat ee ta
A RMN
G13 :
. Ic tex 7
‘ hi - 5 eS : a 7
7 1 : Soe ¥ > ol be Tae qnarrey cont eee Oy

rane<a
Be ~ ; hae a oe ' 7 iA a= wi) fade Ast CrOgr
eu arene Ti tern g A alae= er

7 e q — ap 7)¢€ ie Ws) re an > oJ i w
; esis! te 6S igs £ ony! oa man a cmevenst

Ger gh
wm 2 awa rtd ae i as vg feat a.
‘i - hs Salt cher SA Te er | abate
= .4'3.0! tyes ofp ba eee bee
ne phat
4 el:
po digie 29m) be” ton (emgre ates os
om
= Serpe? wi erty i tha \pecpeonane

é Jlel bron @ = iS bet a am. ‘
neaag ow! nar PRS Say Pore
as, vlad ethtn) Oe a eat
‘a Yeas Ce a er .
uM imn) MeibeT eared ait ys
if? 4 OES, mditiuast “hep bo
re it Tes piv. te 2
* deer EDY Perey on ee
ts Alsat! Prgena8? Sr? ba oe VR
a a ier io
-_ ~
at yy ae
tae Jed 10
iy @/ tate Pad 7
yet! a ; ewe ¢ ’ ae
o ihewu.w> cole Parle

oh@i i 2 wld ad th SD ov
iy hale
Wl etePham ey) aa
a ron pra, a, a gahons
; =! me a ay ae
py : Sat ,egyy oT4
a? oe ie ‘
7 OW opr™virwena ‘a eorndsp Tete,
=? ene, °A 4s [etn OSes
GS) aA
re
4 shew!) 4. eres ©
= i. _Saqpaiat: o Gepgury Pele 9 ey ae a:
am Fa hes ; ieee on WANG 401 Pea.
. aL } ye 1 Hy rab Ve paras py’ oa o ie
;>-ent ' x i.’ oa Gh : WP oouew ye Ogi 2: aca. ime
; age less ail o carted otras Qe" ores! Saes
tn lf om oS Ae CHM Ch dats dale ging
(ae > pa wares Hm ‘2 Wanye, weve6 wh
“6 s% : @ Peres =o ® Rane Vales SS 2)
: fa 4 r g) hy \erere, a Ae rears Be
a P “ J aePerg Rada e705;
a > €e ah a => Seed even Pie at’
SOUND SOURCES 31
Orchestras and chamber music ensembles

Musical instrument acoustics
Musical instrument dynamic range and loudness
String instruments
Woodwind instruments
Brass wind instruments
Percussion instruments and piano
The speaking and singing voice
For the realization of proper concepts in recording, it is a basic requirement that

one have knowledge of the acoustical properties of sound sources, especially those of
the human voice, of musical instruments and of instrumental ensembles. These
properties may be investigated initially without consideration of the sound radiation.
However, it is just those radiating characteristics of the sound sources in a recording
that provide important modifications of the sound properties. The radiating charac-
teristic significance in wind instruments is greater than in string instruments, since
radiation from instrument bells or holes provide a more pronounced radiating behavior.
The radiating characteristics in the near field of an instrument are so significant that
they have a greater influence on the total sound picture than does the choice of a
microphone. The simplest way of selecting the optimum microphone is to place two
microphones at different locations and to compare sounds by switching between them.
The greater importance of proper microphone location@ver proper microphone type)
is most applicable to single microphone recordings and to recordings using support
microphones. /In recordings made at a greater microphone distance, such as is usual
in serious music recording, the influences of the recording space diminish the signi-
ficance of the sound source’s radiating characteristics, so that one can then pay closer
attention to the selection of a suitable microphone.) It is not possible to recommend
specific microphones for specific applications within the purviews of this text, but
general information will provide some guidance in this direction.
32 Sound sources
Instrumental configurations
If more than about 10 instruments form an ensemble, one generally speaks of
an orchestra. The string voices usually account for multiple seats in such an ensemble.
The membership of the symphony orchestra as a larger, and the chamber orchestra
as a smaller, group results from the selection of the composition to be performed. The
size of ensemble generally remains within a range of members indicated by the style
and period of the composition but leaves certain details open. The core of the
orchestra is the string sections, comprising Violin I, Violin II, Viola, Violoncello (or
cello) and Bass. The cello and bass play the same parts for music composed prior to
1800, after which time separate parts were written for each of them. This string
arrangement has remained unchanged till today. What has changed is the size of each
section. As the wind sections were increased during the 19th century, the string
sections had to be enlarged as well to maintain proper tonal balance (fig.A). During
the baroque period, the ensemble was ofttimes enlarged through the addition of 2
oboes and/or flutes and 1 bassoon, sometimes by 2 to 3 trumpets and a pair of timpani
depending on the style of the music. Baroque music always has the so-called figured
bass, consisting of one or two bass instruments (cello, double bass, bassoon) and an
instrument capable of playing chords (harpsichord, organ, lute). After 1750 the wind
sections of the orchestra show steady growth, until around 1900 the orchestra reaches
a size which can no longer be exceeded (fig-A). The double woodwind section in
classical music, aside from some other expansions, has remained standard. It is only
the percussion group which has been expanded in the 20th century.
All serious music groups of no more than 10 soloist voices are called chamber
music ensembles. Instrumental music from before 1600 must be considered chamber
music. We find here neither standards for participating musicians nor other clear
instructions for the playing of compositions. Prior to 1600 it is the wind instruments
which predominate and which were available in much greater variety, but with con-
siderable playing difficulties. Chamber music has developed certain group member-
ships for baroque, classical and romantic music:
Solo sonata: composition for an unaccompanied instrument.

Duo, duo sonata: composition for two instruments. In compositions for
piano and another instrument, the piano is usually not mentioned: a violin
sonata may therefore be either a sonata for violin and piano, or one for unac-
companied violin. A duet is a composition for two melody voices.
Trio: composition for 3 instruments or voices. A frequent form is the piano trio
(violin, piano, cello); a horn trio may be a work for three horns or for one horn
and two other instruments. The trio sonata in baroque music was composed for
four musicians (2 melody, 1 bass, and 1 chord instrument). (Editor: The German
word Terzet refers to a composition for three singing voices.)
Quartet: composition for 4 instruments. One of the most important such groups is the
string quartet (violin I, violin I, viola, violoncello).
Quintet, sextet, septet, octet, nonet, and decet: compositions for 5, 6, 7, 8, 9, and 10
instruments.
In the area of rock, jazz and popular music it is difficult to speak of the instru-
mental configuration in studio productions, because the number of voices or instru-
ments often does not agree with the number of musicians.
A. Standard sym- Instruments Woodwinds ies}pe» n i) Strings

phony orchestra =
ate fe) |
seating develop-
ment (strings 51.68 ” o| B
alf|n /=|s|s
o£ S| & 2 o- Fa 3| 86
=|
number only
02/2 ols| a] °l Sol als] [cl= Q's
approximate)
SISIE//BZ/OlS El El S|S/S Si92
BEC 5; 9} 2].9].9 O]6
; S/2/8a, Qi86 S| 0
Period = ©6|.0/O) O].a; o}.c/ 5/5) 2)s)'5
1750-1770
1770-1790
since 1790
since 1840
since 1870 NM
PO
|
es
Beh
]
horns
~
Sa large
aecuieol &
symphony or-
chestra with
chorus
string instruments
conductor
C. Opera house audience prompter

orchestra pit conductor
opera house orchestra pit stage

Sound sources
o PorintotsWilclcncaisEN OO 2, A 4
BAR CS
Boone violin tI EXER sonia bass
Patera CRA,
American violin | hal violoncello =
;contra-bass
Xs)
icy viola
4E3B++ 3+ eS
German 43- 3+ 3+ Ca [era] ED HED HEB ED
violin | violin I
AQ Stay I
Pec bee eh, violoncello contra-bass
according to Furtwangler
ees
3423+
433+ 3+
“Deep edaed
*HEIHED HEB EB
violin | fins} viola
Bo yea Cor i eee TB ey clarinets bassoons
flutes oboes clarinets bassoons Sh eH, alle ile See oe
E. Seating of woodwind instruments in an orchestra rN |||
flutes oboes
violin Il viola violin I violoncello

violin | violoncello violin | viola
F. Seating arrangement of a string quartet
ae [| i be singer
wines violin etc. violin violoncello
G. Seating arrangement for chamber music with piano

Orchestra and chamber music ensembles 35
This is due to the so-called playback recording process. There are vastly differing
numbers of instruments depending on the style and time of the composition, the
arrangement or the production. Some sort of standard has developed only with a few
musical styles. Basically the requirement is for a more or less elaborate percussion
group, a bass instrument (electronic, string), and an instrument capable of producing
chords (guitar, keyboard). In the most recent decades it is the electric guitar which
has largely determined the sound; often as many as 3 electric guitars are used. Wind
instruments also play an important role (trumpets, trombones, saxophones, clarinets).
In addition such groups also include acoustical or electronic keyboard instruments,
while strings rather form the exception. The big jazz band also has no standard group
configuration, but one can use the following as a guideline: 4 trumpets, 4 trombones,
5 saxophones, 1 clarinet, 1 guitar, piano, bass and drums. Brass bands also have
differing configurations; of first order of importance we find brass (trumpets, trom-
bones, horns, tubas, alto horns, tenor horns, baritone horns), often augmented by
woodwinds (piccolo, flute, clarinet, bassoon). Add to that bass and snare drums,
cymbals, etc.
Seating arrangements
Fig.B shows the stage arrangement for the different instrumental groups of the
symphony orchestra, and fig.D and E the seating within the instrumental sections.
Since the 18th century and until 1945, Germany used the German or classical seating,
while after 1945 theAmerican seating, basically introduced by Leopold Stokowski and >
also suggested in a variation by Wilhelm Furtwangler, became popular. The an
seating is particularly dedicated to precision playing, while the German seating appears
to have advantages in spacial balance, particularly important in stereo recording. In
the opera orchestra the limited space of the orchestra pit determines the seating
(fig.C), which is not as standardized as is the one on stage. Figs.F and G show
examples of the seating of chamber music groups. For rock or popular music the
acoustics of the studio determines the seating (+p.94), while on stage it is mostly
dictated by the visual effects required in show business.
Dynamic range and level

The dynamic range of a large symphony orchestra depends on the dynamics of
the individual instruments on the one hand, and on the composition and the acoustical
properties of the reproducing space on the other. For live concerts with audience, the
background room noise has a direct effect on the softest possible sound pressure level
of the music in the hall: about 35 to 45 dB and only slightly above the room tone.
The highest sound pressure levels are generated by the brass and percussion, lying
seldom above 100 to 110 dB. Thus the dynamic range of a large orchestra may be
between 50 and 75 dB. For works of the 18th century and the commensurately
smaller orchestra, the dynamic range is smaller, about 35-50 dB. Studio recordings
allow a greater dynamic range, since the extreme quiet in the studio permits unmasked
lower sound levels. Studio recordings with over 80 dB dynamics are routinely possible.
The total output level of an orchestra, as perceived by the audience, is dependent
on the number of players, as well as on the particular room, since the diffuse sound
level increases directly as a function of the reverberation time but decreases with
increasing room volume. For a recording, on the other hand, the total level is prin-
cipally a function of the number of players, since the direct sound predominates. The
level of short duration tones, when compared to that of longer duration tones, is largely
room dependent as far as the concert goer is concerned, but is of no practical sig-
nificance in a recording.
36 Sound sources
Musical instrument acoustics
Time analysis of a tone

ical tone is perceived as an organic whole, but acoustically it is compos
of three segments with differing structures which follow one another sequentially:
attack, quasi Salonah condition, and decay (fig.A). While all instrumental notes
naturally egin with the attack, it is only the string and wind instruments which have
a quasi-stationary condition, during which time their acoustical behavior changes but
little. All percussive instruments, to which belong the piano and all plucked instru-
ments such as the harpsichord, transition directly from the attack into the decay
phase. An exception to this is the electric guitar: it can die away so slowly as to give
the impression of a quasi-stationary phase. Wind and string instruments have but a
relatively short decay time. The three-part time behavior of a tone, therefore, cor-
responds to the reaction of a space to a sound event, with reverberation build-up
(attack), quasi-steady state, and decay. The shaping of the sound resulting from the
resonant body of the instrument is modified analogously through influences of the
surrounding room.
# ot
e attack is a significant portion of a tone; it contributes to the recognizability
indies and lasts between 1 and 250 ms, depending on the sound generator
and the attacking system (fig.B). Under 10 ms it has the characteristic of a click
which appears at relatively high level quite separate from the tone itself. Between
a duration of 10 and about 40 ms, the attack cannot be perceived at all; above that
it sounds soft and the build-up of the oscillation becomes clearly audible. The shorter
the attack, the noisier it is. The attack which results from the diagonal editing cut of
a 38.1 cm/s (15 ips), %4" tape creates an artificial attack of 17 ms and is therefore in
the unnoticeable range. Since such edits generally are chosen to come just ahead of
a loud section, the backward masking effect can also serve to cover it.
With decay the level is reduced exponentially as a function of the specific
instrument, just as is the case with reverberation, while the sound spectrum changes
at the same time. The higher harmonics decay faster, making the sound duller with
increasing decay time (fig.C), something which is true for the room reverberation as
well (+p.22). It is easy to see why we use the terminology of reverberation, in view
of the similarity between tonal and reverberation decay. For those stringed keyboard
instruments which are of particular interest because of their decay characteristics, we
obtain decay times of up to 40 seconds at low frequencies, while they reduce to only
a few seconds at higher frequencies. Therefore, the instrument’s own decay time is
quite long when compared to the reverberation decay time. This means that the
echoiness of rooms will not be as distinct for these instruments as it is for wind and
string instruments.
Quasi steady-state oscillation characteristics

The quasi steady-state for instruments, with the exception of electronic ones, is
never an even one, but rather displays typically regular and also especially irregular
unsteadinesses of the most important attributes of sound. The tremolo refers to a
strong amplitude modulation of the tone, which, for string instruments, is produced
by a rapid oscillation of the bow against the string and in wind instruments by the
so-called flutter-tongue. In its pure form the vibrato in actuality is a frequency
modulation but in musical instruments it is almost always combined with some ampli-
tude modulation. The vibrato is used by all string and wind instruments (with the
exception of the clarinet and french horn) to improve the tonal quality.
Musical instrument acoustics Sa
A. Sections of a level
musical tone
attack quasi-stationary decay time

| condition
B. Attack time of tone start

musical .
instruments PR ONSEN,
normal soft Ashe hurr Sespooeguee time (ms)
attackera
@)
Bi Sine
50 400)
100 WACO,
150 S20 oleae
250 ee
remarks:
percussion extremely short: castanets
(1 ms), triangle (4 ms)
rel. long: timpani (to 18 ms)
plucked string extremely short: string
piano
wood (hard attack) flute, low register to 120 ms
brass (hard attack) horn, low register to 80 ms
wood (soft attack) flute, low register to 180 ms

oboe, low register to 120 ms
string (hard attack) contra-bass, low register
to 110 ms
trumpet, low register
brass (soft attack) to 180 ms
string (soft attack) contra-bass, low register

to 450 ms
C. Decay of musical
instruments (piano)
38 Sound sources
dB $ level
D. Vibration form
and spectrum ' 40
sine 30
pulse sequence
wave 20 frequency
10 :
harmonic
triangular
wave
1. 3.5.7.9.1113.15. harmonic
sawtooth
wave
WESoe ~ Oy harmonic
square
wave || || |
1-3) S AONB 15. “harmonic
impulse pelle al
E. Formants
oboe
| \ |
| |
- i | ! !
woodwind tia |
instruments
brass wind
instruments ) french horn :
]
\ }
A
1 |
7
string j
viola | |
instruments |
i}
|
i}
\ I
ee
for comparison:
| frequency
the pure vowels
1000 2000 3000 4000 Hz
Musical instrument acoustics 39
All wind instruments including the organ, and all string instruments generate
musical tones (sounds) which are made up of so-called harmonics (partials). These
partials have a harmonic relationship, i.e., the frequencies of the harmonics are a
whole integer multiple of the frequency of the first harmonic or fundamental. Devia-
tions from the regularity of this spectrum are found particularly in the piano sound.
The reason is the thickness and stiffness of the strings. Oscillating plates or pipes such
as cymbals, tom-tom, gong or bells, and oscillating membranes such as kettle drums
(timpani) and all types of drums, but also rods such as triangle, display a more or less
non-harmonic spectrum. They mask the sensation of tones as in bells, gong, triangle
or timpani, or give the sound an entirely noise-like effect as with cymbals, tom-tom
and in drums. (thefrequency range of the spectrum not only depends on the type of
instrument and on the location of the microphone or listener, but also on its place-
ment within the dynamic range}(+p. 40). The frequency range may increase by a facto Te
of three to ten’in the transition from pianissimo
to fortissimo. The lowest register Jo 51
instruments, with fundamentals starting at about 25 Hz are the double bass, con- ovi
tra-bassoon, bass tuba, organ, bass drum etc. The usual bass instruments range begins
at about 60 Hz. The upper boundary of the spectrum for double bass, contra-bassoon,
bass tuba and timpani is at about 5000 Hz; for most other instruments between 10 and
15 kHz while some percussion instruments such as the triangle go beyond that (+p.31).
The wave form-—fig.D shows some idealized basic wave forms—determines the
intensity and frequency range of the harmonics, i.e., the spectrum envelope. It also
determines whether all harmonics or only the odd ones are present. The idealized
wave forms shown in fig.D only exist in electronic instruments, but natural wave forms
display enough similarities to permit a derivation of the natural wave forms. The
flute comes closest to the sine wave form, especially when playing softly, while the
string instruments follow more the sawtooth wave form. Clarinets are most similar
to square wave forms while double reeds and, to a minor extent, the brass instru-
ments display a wave form closer to pulses.
Formants are resonance-like, amplified harmonics which have a fixed position
in the spectrum regardless of the tone being played. Formants are particularly an
attribute of string instruments, double reeds and brass (fig.E). They are also respon-
sible for the differences in speech vowels (fig.E). It is for this reason that the formants
impart a kind of vowely quality to the instrument sound. The bright, open vowel
sound "ah" is to be found in the violin, trumpet and oboe. The bassoon sound is best
described by the "oh" formant. The so-called nasal formant is located in the range of
1800 to 2000 Hz and is highly developed in the saxophone.
The background noise is as much a part of the sound as are the harmonics. It
is strongest in string instruments and displays the resonant quality of the particular
instrument. Somewhat less, but nevertheless quite indicative, is the noise component
of the woodwinds. The flute is accompanied by a rather typical breathiness while the
noise components are weakest in the brass. These noise components are found at a
level of about 30 to 50 dB below the level of the strongest harmonics. The noise
modulates the amplitude behavior of the sound which thereby undulates by several
dB. A sound’s noise components are much more important in recording than for the
live listener since microphones are much closer to the instruments than listeners are.
They increase the instruments’ presence and intensify the sound. Breathiness in wind
instruments identifies an individual player.
40 Sound sources
Musical instrument dynamic range and loudness
Technical dynamics
y In electro-acoustics the dynami is generally defined as the level range
between an upper and lower fh ide alogously we have the dynamic range
of musical instruments and similar sound sources as the difference between the
highest and the lowest producible sound levels} When the room tone is included,
something one can hardly separate from the instrument sound in practice, the lower
boundary level is hardly definable because the sound level, together with the rever-
_ beration decay, dips into the unavoidable background noise. fetechnical dynamic
range jis firstly made up of the microphone dynamic range which lies between the
: “overload point and the self noise level. [System dynamic rangeJdescribes the dynamic
values of the entire technical transmission chain. One must differentiate between the
maximum and the effective system dynamic range which describes the true usable level
range considering the overload reserve and a reasonable number of dB from the base
noise. One of the problems in quoting the technical dynamic range is the fact that
there are several standard methods of noise measurement which may yield differences
of up to 10 dB. When measuring the highest usable level in analog transmission or
storage, it is the peak program level meter (PPM) displaying quasi-peak levels which
is the most suitable instrument, while for digital transmission or storage one should
use a real-peak level indication which may yield up to 10 dB higher peak levels.
When using the usual PPM for digitally processed signals, one must allow for an
overload reserve. The program dynamic range is the dynamic range which is estab-
lished, considering the desired or possible playback dynamic range (fig.A.)
Level and loudness

The levels which define the dynamic range are physical magnitudes which are
important to the audio field. However, during the enjoyment of music and speech the
physical magnitudes are subjectively evaluated as well by utilizing our prior audio
experience. The values of dynamics, therefore, only provides a first, rough estimate
about the possible differences between sensed loudnesses. Loudness as a measure for
the subjectively sensed sound intensity is only defined for sine waves. This is done
using contours of equal loudness and finding the sound pressure level of the equal
loudness sine wave tone at 1000 Hz. This is also the loudness level in phon. By
means of the loudness curve, this value is then converted to loudness in sones (fig.C).
Non sine wave signals may be converted to loudness values by loudness comparisons
with sine wave signals or through application of certain standardized calculations {In
general, a level increase of about 10 dB corresponds to a doubling of the loudness; }
a doubling of the sOund pressure or tfie voltage leads to an increase of only 6 dB.
Only a tripling of the sound pressure ao a multiplication of the sound power by
ten doubles the loudness. An increase in the number of instruments playing at equal
volume, for example violins, leads to the following level increases: twice the number
of players provides 3 dB more; four times the number provides 6 dB; eight times 9
dB, ten times 10 dB, and sixteen times results in a rise of 12 dB in level. This shows
that for a significant increase in loudness, one must provide an inordinately greater
number of musicians. fthe level relationships are exactly reversed when several
microphones with about equal output level are to be mixed together. If two such
microphones are mixed then the level of each one must be 3 dB below the desired
total level, for 4 microphones it would be 6 dB and so forth.)
Musical instrument dynamic range and loudness 41
recording microphone sound transmission loudspeaker

space
Ors : N\
' > headroom
: SPIeimite acta, Gel eee ed. Pens aon iat ie wr -%

ff + (ca. 130 dB)
program reproduce
dynamic range usable sates dynamic
of musical : system maximum aah dB) range (ca.
instruments Jes dynamic > system 35-55 dB)
| | Pete. phone | range dynamic
S dynamic range
= equivalent ge
SPL (A-wtd
ca. 20 dB)
Ppp+r< ) guard area

- > to noise
| p reverb and 1 micro- floor
| | room tone H phone self-t equipment
! noise it noise
A. Dynamic range, concepts
100 100!phon
0 me
:
2- 60 ee
: 2
=
2 40 3
5 20 20 Ke)
i he a
ZOE SO MOON sOO0T 1s 2 5) Kz. 20

frequency ——= loudness level —-=
B. Equal loudness contours for C. Relationship between loudness

frontally impinging sound frontally impinging sound
soft Fae er ree :

ppp pianissimo possibile as soft as possible
pp pianissimo very soft
p piano oe if
mp mezzo piano half so t :
mf mezzo forte half loud identical
f forte loud
ff fortissimo very loud
loud fff forte fortissimo as loud as possible
D. Dynamic steps in music

wie
Sound sources
violin
viola
violoncello pLSet ee, EL
contra-bass hos SRP IPL oxo
. Ma
flute 100 200 500 1000 2000 Hz
oboe
violin
clarinet
tenor saxophone
bassoon
french horn
trumpet
trombone
tuba
100 200 500 1000 2000 Hz
timpani trumpet
bass drum
snare drum
cymbals
soprano
bom i
alto A
tenor
bass (Sas See 100 200 500 1000 2000 Hz
recorder
harpsichord
organ
guitar
orchestra
dB
Sao 70 dB = 100 200. 500 1000 2000Hz
level = clarinet
E. The dynamic range of instru- F. The pitch dependence of dynamic range

ments, voice and orchestra
level
G. The transition from

pianissimo to
fortissimo of
a tone of in-
creasing loud-
ness (clarinet)
Musical instrument dynamic range and loudness 43
Two further factors influence the relationship between(level jand@oudness:) (1)

for sound components below 500 and above 5000 Hz the ear is less sensitive; and (2)
for program material with like meter indications, sound level patterns with numerous
spikes appear to be softer than those without spikes.
Level and timbre dynamics

The dynamic range of a musical instrument or ensemble is categorized in steps
of musical dynamics. They reach from the lowest playable to the highest playable
loudness (fig.D). The individual dynamic steps are differentiated in their loudness or
acoustical level (level dynamics) and in their sound coloration or spectrum (spectrum
dynamics). The level dynamics describes only the level differences. Between two
wwe of the tainty, ee there usually exists a 6-10 dB level difference. The
ound leve ences in live music are dependent on several factors{on the
Bound feel of the wee source\(a function of the particular instrument and playing
technique), @n the distance, onthe reverberation time ,Jand on (the room size.) As a
result, the sound level values may vary within wide limits. Of critical importance,
especially for the listener (less for the recording) are the sound levels of those instru-
ments played in large rooms which are too weak, since their low frequency sound
components sound even lower due to the sensitivity losses of human hearing. Fig.E
shows the dynamic range of musical instruments, the dynamic range of the human
voice, and that of a large symphony orchestra under actual conditions. On average,
the string instrument level is about 10 dB below the level of the wagd winds while
these are 10 dB below the level of the br DTgss. For an orchestra this is equalized
through commensurate membership of the various instrument sections to achieve a
balanced sound picture. For many instruments the dynamic range is a function of the
register in which they play. Fig.F shows typical dependencies. For strings, but also
for piano, guitar and harp, the dynamics and absolute levels are fairly close over the
entire range. For the brass the dynamic range in the upper register becomes smaller
while the absolute level increases. With flutes, and especially recorders, the dynamic
range is evenly small while the level increases withiincreasing frequency. The clarinet,
by comparison, has a rather large dynamic range in its mid-range. These acoustical
instrument phenomena must be considered by the experienced composer or arranger.
The timbre or spectral dynamics are just as important in music and recording
techniques as are the level dynamics. Because of the fact that every dynamic step is
characterized by a specific spectrum, it may be recognized independently of the
audible level. This is a basic prerequisite for the performance of music altogether
because a forte passage at a close listening distance will never become piano at a
greater distance. [The number and intensity of the harmonics increases with loudness.)
The formants, too, are built up one after the other. This loudness defined by the
spectrum is therefore in reality not a true loudness, but rather information about how
loudly the instrument was played. It is this very spectral information which makes it
possible to alter the level for technical reasons altogether, i.e., to reduce the dynamic
range. Going too far with such dynamics restriction causes contradictions between
spectral and level dynamics which are not acceptable. Fig.G gives an example of how
the spectrum of a swelling tone changes for a transition from piano to forte.
The ear normally judges loudness by utilizing the distances of sound sources.
Therefore it does not determine whether two sound sources are equally loud to the
listener at his listening position but whether they are truly equally loud. The ability
to judge loudness, of course, is meaningful for our orientation in the world around us.
It also leads to the fact that the sound pressures at the ear when listening with
headphones is considerably higher than that from loudspeakers for a sensation of
equal loudness.
44 Sound sources
String instruments
Instruments
There are four prevalent string instruments: violin, viola, violoncello (or cello),
and contra bass (or double bass). The violin, viola and cello differ mainly through
their size but only minimally through their shape. The double bass is built differently
in several details. Besides these, we occasionally use historical string instruments such
as the viol or gamba which is played on or between the knees, and from this group
of instruments especially the tenor gamba which is most like the cello in its pitch range
and size.
Application
In a symphony or chamber orchestra, the string instruments are choral, meaning
that there are several instruments in each section. There are traditionally five voices:
violin I, violin II, viola, cello and bass. The membership of these sections in large
orchestras and in the order given is 16-24, 14-20, 12-16, 10-14 and 8-10, and in small
orchestras 8-10, 6-8, 4-6, 4-6, and 3-4. These groups of string players form the core
of the orchestra. In chamber music they form the most popular instruments along
with the piano, but here each instrument (except for the violin) usually appears only
once. There are relatively few compositions for a single, unaccompanied string
instrument. In jazz the normally plucked bass is of great importance, and recent folk
music influences have also returned the violin to this art form. In big bands, the
strings serve about the same function as in a classical orchestra. And in popular
music strings are used for background effects or as soloists; however, they serve a
subordinate role here and, except for their use for solos, are often replaced by elec-
tronic strings which either synthesize the string sound or have memory circuitry
containing the sounds of real strings.
Sound acoustics
Sound generation: Because the string sticks to the rosined hairs of the bow and
then periodically rebounds when its excursion becomes too large, we obtain sawtooth
like string vibrations. These contain the total sequence of harmonic partials at high
amplitude_as their acoustic raw material which abates with increasing frequency
(+p.36). |The vibrations are transmitted through the bridge to the resonance body
which, in turn, radiates them to the environment. This resonance box is part of a very
complex resonant system which greatly modifies the original form of vibration. To
shape timbre_and loudness, the player may change the speed, pressure and position
of the bow. [The intensity is really only influenced by the speed of the bow/while fits
pressure and position on the string determine the timbré)(fig.A).
Attack: Among all of the instruments the strings have the longest duration attack
phase. For normal playing, the tone is fully formed only after 100 ms (for the double
bass after 400 ms). For a sharp down beat, these times are reduced to 30-60 ms and
150 ms respectively. This may explain why one often hears the comment that the bass
voice appears somewhat delayed. Pizzicato tones have a much shorter attack of under
20 ms and therefore give the impression of greater precision than bowed tones.
Tonal range (fig.B): The tonal range is limited at the bottom by the tuning of
the lowest string while the upper boundary is strictly a function of the musician’s skill.
Frequency range (fig.B): In string instruments the sound partials beyond 10 kHz
are relatively weak. The individual frequency range is highly dependent on the playing
method. [The response range widens towards the high end with increasing bow pressure
7
String instruments 45
pure’ M
3 dB
mee
Qo H8 oO
8
wD 2 &
::
0)
Se
a5
So
122
E
g rough ES 0
> unstable ) 100 cm/s 150
S bow spesd— —=
unstable
w/o fundamental
bow pressure
lara
25 10 end of 5 string length
finger board
bridge string finger board
A. The influence of playing technique on the sound of string instruments
frequency ranges
of the spectra ; violin
and formants
| viola
i Va Sale
BS a lel einyorellfe
contra-bass
500, 1000 2000 gva_, 5000 10000 Hz

ao
contra-bass
l
tonal ranges violoncello

viola
violin
B. Frequency ranges of the spectra with formant positions and tonal ranges
46 Sound sources
UYze
A
1000 — 1250 Hz
Gi
Lx Y NSS
aesrewils ae ard res eh 1 i ‘ eS

oS on S NO 0.3 0,4 0,50,6 0,8 ks. 2 3 J
level
sound
———.»
40 ~
\ source ca el ore
Terei d Ue eee er em | PETN feeoe
ha eT eperal b.
ONS We Wier WAN ONs (is; os sh 3 4.) 95.65 Khiz 0
D. Frequency response of a violin Treensy

a. in the near field at a particular microphone location
b. In the near field for super imposition of several frequency response
curves of various microphones.
String instruments 47
and decreasing distance of the playing location to the bridge (fig.A).

Formants (fig-B): The resonant properties of the sound box, its form, dimensions,
materials, and its construction determine the frequency position of the formants, i.e.,
the areas of accentuated sound components. The resonance areas are fixed while the
frequencies of the played tones change constantly, so that constantly changing partials
fall within the range of the resonances. As a result, the timbre may vary greatly from
tone to tone. emcae soon insen's have no sharp resonances and therefore
have a unifie
. °
all
r the wie It is the formant in the vicinity of 1000 Hz —sny
which gives the violin its characteristic Pint. It provides the bright, open character-
istic which is found also in the vowel "ah" and which makes it the preferred vowel for
singing (la, la, la). The nasal formant between 1500 and 2000 Hz gives the viola its
characteristic sound. The cello also has a certain nasal character but is most signifi-
cantly described by the width of its formant range of between 2000-3000 Hz, which
lends the instrument a certain sharpness which allows the cello to sound brighter at
higher frequencies than the violin in spite of its larger size.
Noise components: Typical is the relatively strong attack noise which, for wind
instruments, may be 20 to 30 dB louder than the wind tones (with the exception of
the flute). This noise has a continuous spectrum which mirrors the resonant properties
of the instrument identically with every tone. Of special note is the buzzing sound of
the double bass which lends it its special sound within the orchestra and which is
caused by the vibration of the bow hairs. The noise components of the string instru-
ments are largely independent of the playing intensity, i.e., for softly played tones they
are relatively the loudest.
Dynamic range and level: The dynamic range is relatively even over the entire
tonal range (~p.40). It only gets a bit restrictive towards the upper end of the frequen-
cy spectrum. The sound levels are quite low when compared to the wind instruments;
on average 10 dB lower than the woodwinds and 20 dB lower than the brass. These
differences are compensated by choice of the number of players in each section
(+p.40).
Radiating characteristics
The radiating characteristics are basically caused by vibrating wooden sections
of the resonance or sound box at various amplitudes and in various phase relation-
ships. This results in relatively complex behavior which varies within certain limits
(fig.C). The radiating pattern in the frequency area of the formants is restricted to
a small angle while at low frequencies we find an omni-directional radiating pattern.
Towards the upper end of the response range, the radiating pattern is not narrow like
the brass instruments’ if one examines a wider response range. The frequency response
for a microphone position at close range shows something of a comb filter in its
micro-structure which may lead to an unnatural edginess. It is only the superim-
position of the response curves from all directions—something best accomplished by
reverberation—which smooths out the frequency response curve (fig.D). Therefore,
it is recommends rat lu hones be placed at a greater distance from strings,
cially for serious music. The basic difference between natural and artificial
reverberation is especially pronounced in string instruments: while the natural rever-
beration represents an integration of all the radiating directions of the instrument,
the artificial reverberation imparts the specific frequency response at the location of
the microphone.
48 Sound sources
Instruments
The woodwind instrument section encompasses a relatively large number of
different instruments which differ greatly in the way they produce their tones. Not all
of them are made of wood. The flute is made of a silver amalgam and the saxo-
phone was made of metal right from the start. The most important woodwinds today
are the flute, the oboe, the clarinet (in Bb, and more rarely in C or A), the bassoon,
the english horn and the saxophone. Add to that those instruments derived from the
main ones: from the flute family the piccolo and alto flute; from the oboe group the
oboe d’amore; from the clarinet family the small clarinet, the basset horn and the bass
clarinet, the contra bassoon, and from the saxophone family the alto, tenor, soprano,
baritone and bass saxophones. Historical woodwind instruments for baroque music
are the transverse flute which is the predecessor of the modern flute, the various
recorders, and baroque oboes and bassoons. The instruments of the baroque era
contain a multitude of recorders and other flutes, but especially double pipe instru-
ments of different constructions and timbres.
Most woodwinds are written in a tonality different from the one which is heard.
In other words they transpose.(fig.A).
Application
In the classical symphony orchestra we normally find two each of flutes, oboes,
clarinets and bassoons. For music prior to 1800 their number is reduced. In the 19th
century their number has gradually increased through addition of the piccolo flute and
contra-bassoon, then the english horn and the bass clarinet (+p.32). Chamber music
with exclusively wind membership stems solely from the 18th century. Strings were
added after that. Generally speaking woodwinds are not as important as strings in
chamber music. In pop and jazz the saxophone and clarinet have found popularity
with the flute coming along as well. The rest of the woodwinds find only occasional
use in popular music, but almost always in semi-classical music. Brass bands partly
use woodwinds as well, especially clarinets, flutes, bassoons and saxophones.
Sound acoustics
“= Sound generation: Woodwinds are divided into three groups according to their
embouchure: theso-called vibrating air reed of the flute aims at a sharp edge and
oscillates back and forth between the resonating air column within and the surround-
ing air without. As a result one hears a fairly constant embouchure noise which is
typical for the flute sound. Clarinets and saxophones have a simple bamboo reed,
which alternately opens and closes the instrument under the embouchure and lip
pressure of the artist thereby exciting the air column in the resonant tube to oscilla-
tion. With oboes and bassoons this same function is performed by a double bamboo
reed. The quality and suitability of the double reed is a constant problem for the oboe
player—far more than with the single reed. Because of this, it is a common practice
for oboists and bassoonists to make their own double reeds, i.e., varying the hardness
or easiness in the embouchure considering the type and tonal range of the music to
be played. Clarinet players, on the other hand, can make do with manufactured
reeds.
Attack: With the exception of the flute, the attack time of 10-40 ms is con-
siderably shorter than that of the strings and, therefore, more precise. By contrast
Wood wind instruments 49
Instrument Notation Pitch referred ak |

to notation
flutes concert flute G-clef as written

piccolo (small flute) G-clef one octave higher
alto flute in G (in F) G-clef a fourth (fifth) lower
soprano flute G-clef one octave higher
alto recorder G-clef as written
oboes oboe | @clef as written
english horn G-clef a fifth lower
oboe d’amore |G-clef a minor third lower
aaa clarinet in B G-clef a major second lower (as

(in C; in A) written; a minor third lower)
small clarinet in Eb G-clef a minor third (major second)
(in D) higher
basset horn in F(in Eb) G-clef a fifth (major sixth) lower
bass clarinet in B | bass clef a major second lower
bassoons_ | bassoon bass clef as written
contra-bassoon _|bass clef one octave lower
saxo soprano sax in B G-clef a major second lower
phone alto sax in Eb G-clef a major sixth lower
tenor sax in B G-clef a major ninth lower
baritone sax in Eb G-clef one cotave + major sixth lower
bass sax in B hess two octaves + a major
second lower
Eetoretest
he oOo
A. Woodwind instrument notations
spectrum and formant

response ranges
a bo bo hee
=
giana |
contra-bassoon
{=
bassoon
L
clarinet
L
oboe
l
concent flute
tonal range l
piccolo
B. Spectrum response ranges with formant positions and tonal ranges

_—— 7
50 Sound sources
low sound partials

(fundamental heavy; tubby)
high sound partials

(sound bright; edgy; tight)
C. Main radiating areas
flute
ey
250 — 600 Hz 3000 Hz 8000 Hz
clarinet
(oboe similar)
ESS
FOO
BESSOOOO
SSRIRES
OOOO
bassoon
2000 Hz 5000 Hz
D. Radiating characteristics
Woodwind instruments Sil
the flute, in its lower range, may display an attack time of over 150 ms and, therefore,
its onset will appear commensurately soft.
Tonal range (fig.B): The tonal range encompasses two to three octaves, also
dependent on the skill of the player. The contra-bassoon is the orchestra’s lowest
instrument and the piccolo its highest.
Frequency range (fig.A): The spectrum of the woodwinds reaches generally to
10,000 Hz. The flute has a distinctly smaller range of only about 6000 Hz. Of all the
instruments of the orchestra it is the flute which has the strongest fundamental. It
sounds sine wavy. The other instruments have strong 2nd and higher harmonics.
Formants (fig.B): Among the orchestral instruments, the double reeds are most
strongly characterized by their formants which form during the embouchure due to
the special form of oscillation of the double reed. The oboe has the same formants
as the French nasal vowel sound "in". Its sound is therefore nasal and bright. For the
bassoon the vowel sound "aw" is characteristic. The flute’s weak formants give only
little information about the difference between individual instruments, while in the
clarinet they also have little meaning. The flute is largely recognizable by its em-
bouchure noise and the relatively small amount of harmonics, the clarinet by its
suppressed even order harmonics levels.
Noise components: The background noise is relatively weak when compared to
the string instruments. The flute forms an exception with its typically flute-like em-
bouchure noise.
Dynamic range and level: A strong pitch dependence of the dynamic range is
typical for the woodwinds. Therefore even adjacent tones may have differing dynamic
behavior. The flute displays only limited dynamics in its upper register. The oboe has
this in the lower register. The clarinet has a very wide dynamic range in its mid-range.
The level of the woodwinds is about 10 dB higher when compared to that of the
strings, meaning that they sound about twice as loud. The levels increase somewhat
with increasing pitch, especially for the flute (~p.40).
. Radiating characteristic
a Summarized simply, the lower and mid-range tones up to about 2000 Hz, emanate
sideways from the finger holes of the instruments. Higher pitched ones beginning at
about 3000 to 4000 Hz radiate from the bell (fig.C and D). The tonal timbre changes
more severely with the radiating direction than is the case with the string instruments.
It is therefore more important to select the proper microphone position rather
than the most suitable microphone type. This selection becomes even more decisive
the closer the microphone gets to the instrument. With increasing instrument-to-micro-
phone distance, the diffuse sound merges all of the radiating directions into a total
sound, which becomes ever more independent of the microphone position. A close
miking position absolutely requires that the instrument always be held in a fixed
position, something of which normally only the best studio musicians are capable.
The flute acts like an acoustical dipole since it also radiates sound from the
mouthpiece. This causes certain sound cancellations within narrow angles, which, in
turn, makes it necessary {that a microphone be placed in such a position that it is
equidistant from both ends of the instrument.) By contrast to the other woodwinds,
the flute emits audible breathiness from its mouthpiece.
In a saxophone the radiation from the finger holes and from the bell coincide
because the bell, with the exception of the soprano sax, is directed upwards. This is
also true for the bass clarinet.
52 Sound sources
Instruments and their application

The following brass instruments are found in a symphony orchestra depending
on the style and era of the music: trumpet (2 or 3), french horn (2-4), trombone (3),
and tuba (1), more rarely cornet, natural horns and so-called Wagner tubas (~p.32).
In the mixed make-up of classical chamber music written for strings and woodwinds,
we usually find only the french horn from the brass family. However there are several
types of ensembles consisting only of the brass instruments. In brass bands (usually
military marching bands), we find trumpets, french horns and trombones, as well as
cornets, alto horn, tenor horn, baritone or euphonium and tubas in their helikon or
Sousaphone structural versions. Concert bands consist of a woodwind and brass
mixture. In jazz and pop music, trumpets and trombones predominate, and in the
traditional jazz styles such as dixieland, one also finds cornets, helikons and Sousa-
phones. The trumpet used in jazz is particularly narrow and short. Among the
historical brass instruments, there is the valve-less trumpet and horn (natural trumpet
and horn), the trombone and the cornetto (German: Hélzernen Zinken). Most
instruments are produced in different forms and sizes.
Trumpets, french horns and other horns are so-called transposing instruments
which means that the played tones deviate from the written notes. The following are
non-transposing: the trumpet in C, the french horn in C, and the trombones and tubas.
Sound acoustics
Sound generation: The brass instruments’ mouthpiece serves as a support for the
lips which, similar to the double reed, periodically interrupt the air stream. The
interrupting frequency primarily depends on the resonant frequency of the air column
within the instrument. The pitch is not influenced by finger holes at the side, as in
the woodwinds, but by the tension of the lips, the air pressure and the extension of
the instrument’s length due to interchangeable, insertable tubing by means of valves
or, as in the trombone, by means of a telescoping, u-shaped slide. The result is that,
by contrast to the woodwinds, the entire sound emerges from the bell. This fact makes
it possible to give an acoustical function to the brass bell which it cannot have in
woodwinds, because there it only gives off the higher frequency components of each
tone. In view of the fact that the bell of the brass instrument radiates all of the sound
components, it is possible to optimize their sound energy transmission to the room.
This increases the sound intensity of these instruments, which made them so useful
as signal and fanfare instruments out-of-doors. (The bell acts as an acoustical trans-
former) and matches the low acoustical source impedance of the instrument to the
higher acoustical terminating impedance of the room, thereby significantly raising the
efficiency. A further improvement is provided by the great directional effect of the
brass bell.
Attack: For a soft embouchure, the attack takes between 40 and 120 ms; for the
trumpet up to 180 ms; for a sharp embouchure 20 to 40 ms; and for the french horn
up to 80 ms. Typical for the attack of brass instruments is the so-called chiff which
lasts on the order of 20 ms which contains predominantly harmonic contents below
1000 Hz. Too strong a chiff with an otherwise soft attack produces the well known
squeal which is difficult to avoid, especially in french horns and historical brass
instruments.
Instrument Notation Pitch referred

to notation
Trumpets trumpetin Bb (in C) G-clef a major second lower (as noted)
(Instruments played today; player’s choice.)
trumpet in G (in F,in G-clef a fifth (fourth, major third, minor
E in Eb in D, etc.) third, major second) higher
(Traditional instruments, but still notated that way today.)
bass trumpet in Eb = G-clef a major sixth octave, major ninth)
(in C, in Bb) lower
Horns horn in F G-clef, a fifth lower
bass-clef
(low parts)
(The instrument played today, also combined with the Bb alto horn.)
horn in C alto (in B, G-clef as noted (a minor second, major
in Bb, in A etc.) second, minor third etc. lower)
horn in C G-clef one octave lower
(Traditional instruments, but still notated that way today.)
Trombones tenor trombone, tenor-clef, as written
tenor-bass tromb. bass-clef
bass trombone bass-clef, as written
tenor-clef
Tuba bass tuba bass-clef as written
Cornet cornet in Bb (inC) G-clef major second lower (as written)
Small flugelhorn G-clef a major second lower
horns alto horn in F (in Eb) G-clef a fifth (major sixth) lower
tenor horn G-clef a major ninth lower
baritone (Euphonium) bass-clef, as written
tenor-clef
G-clef, a major ninth lower
(brass band
music)
A. Notation of the brass wind instruments
frequency range
of the spectra
and formants
1000 2000 5000 10000 Hz
bass tuba
horn
trombone
tonal ranges
trumpet
B. Spectrum response ranges with formant positions and tonal ranges

oe
54 Sound sources
low sound sound increas-

partials ingly brighter
(sound dull)
high harmonics
=
(sound bright, sharp)
Sao |
sound increas-
low sound ingly brighter
partials
(sound dull)
C. Main radiating areas
trumpet
(similarly trombone)
1000 Hz 1500 — 2500 Hz 4000 — 15000 Hz
french horn
1000 — 1300 Hz 2000 — 3000 Hz 3000 — 5000 Hz
D. Radiating characteristics
Brass wind instruments mE,
Pitch range (fig.B): The upper boundary of the pitch range is largely dependent
on the skill of the player, just as in the string instruments. In the orchestra, the
individual players specialize by assignment to a particular voice (e.g., Ist, 2nd, 3rd and
4th french horn); the higher voice to the 1st or 3rd, the lower to the 2nd or 4th horn.
Frequency range (fig.B): The widest tonal range occurs when playing fortissimo
and depends on the pitch range of the particular instrument. The trumpet, as the
highest pitch instrument, is capable of frequency components to 15,000 Hz; the french
horn to 10,000 Hz; the trombone to 7000 Hz; while the lowest pitch instrument, the
tuba, reaches but to 2500 Hz.
Formants (fig.B): The level increases in the formant regions are not as clear
and in their frequency range not as even as is the case with the double reed instru-
ments. Therefore their influence on the sound characteristic is lower.
Noise components: Noise content in brass instruments hardly play any role at all
due to their very low levels.
Dynamic range and level: The dynamic range displays a pitch dependence
(+p.40). For the trumpet the dynamic range gradually reduces from about 30 dB at
the low frequency end to about 10 dB at the highest pitch tones. The french horn has
a wide dynamic range of about 40 dB in its mid-range, reducing to about 20 dB at the
upper pitch range. The trombone has the highest dynamic range among the brass
instruments with values to 45 dB for certain tones. Brass instruments, therefore, have
a rather uneven dynamic behavior. The level of low or high pitch tones, as with the
dynamic range, is rather pitch dependent. For softly played instruments it increases
by about 30 dB with increasing pitch. This means that low notes can be played very
softly while high pitched ones can be played only loudly. The brass instruments are
the loudest of the orchestra if one ignores certain percussion instruments. They are
on average 5 to 10 dB louder than string instruments. Therefore it is not surprising
that their number in the orchestra is smaller, there being some 60 strings but only 10
brass players in a large orchestra. The brass nevertheless can outplay the strings while
the trombone achieves the highest levels.
Radiating characteristic
By contrast to the woodwinds, the directional characteristic of the brass is largely
rotationally symmetrical about the bell and is therefore easier to manage. T'the
radiating pattern becomes narrower with increasing pitch. The result is that the sound
of a brass instrument becomes duller the farther off axis one moves)(fig.C). This
narrowing of the radiating pattern is not entirely regular. For the trombone, it widens
once more at about 6000 Hz, for the trumpet at 800 Hz. While the principal radiating
pattern for the trugypet and trombone points forward, the french horn directs it
towards the back as a result of the way the instrument is held. The tuba directs its
sound upwards. The directional pattern of the french horn is more complex than for
*
the other instramedtanmh is i divided into several angles (fig.D). The very strong
directionality of the brass instruments, especially of the trumpet and trombone, allows
the sound level to attenuate much more slowly in the direction of maximum radiation
than in undirected other directions. This increases the reverberation radius for these
instruments significantly (+p.26). The result is that, at greater distance from the bell,
these instruments assume less of the room characteristics than do other instruments.
Furthermore, the increased reverberation radius displaces the orchestral balances in
favor of the brass with increasing distance, both as to instrument sound level and
presence.
56 Sound sources
Percussion instruments and piano
Instruments and their application

The number of percussionists in a symphony orchestra is relatively small when
compared to strings and winds. For compositions from before 1800 there usually are
only two timpani, in the 19th century often four or more with added bass and snare
drums, cymbals, triangle, tom-tom, etc. In the 20th century the number, variety, and
meaning of percussion instruments has been significantly enlarged. Added have been
percussion instruments stemming especially from folk music and African, South and
Central American and Asian countries. Some entered through the jazz world, others
came directly into the orchestra.
In pop and jazz, the percussion group, also called the battery, has a very basic
function. By contrast to classical music, it is constantly
in action (beat). The standard
outfit comprises at least one each of a bass drum, snare drum, at least two cymbals
of different sizes, a foot operated cymbal pair (hi-hat), as well as two small tom-toms
and one or more large tom-toms; add to that instruments depending on the style, such
as bongos, congas, maracas and guiro, cow bells, wood blocks, gongs, etc. The piano
is today’s most universal instrument, appearing in all types of music both as soloist
and in ensemble.
Sound acoustics
The sound of percussion instruments when compared to other instruments is more
noise-like. This is especially true for the drums. Or it is marked by a certain out-
of-focus tonality, caused by non-harmonic spectra as, for instance, for the triangle,
gong and bells. The attack is extremely short because it is struck and this gives the
battery its distinctive sound. Aside from the timpani and bass drum, the battery’s
spectrum extends to very high sound partials, often beyond the audible or reproducible
range of 15,000 Hz (fig.A and B).
The fir
impani have a precisely perceivable pitch as the only such percussion
instruments and are, therefore, written as music in the score. They are usually set
to the required pitch by means of a pedal after first having had their membrane
tuned. Timpani of different diameters are assigned different pitch ranges. The
response only reaches to about 2000 Hz and the timpani, therefore, lack upper har-
monics. The type of stick used (felt, foam, wood) determines the spectrum. Both
the dynamic range and level of the timpani are extremely high.
The big drum, in pop and jazz called the bass drum, has a very pronounced low
frequency spectrum reaching, as it does, only to about 5000-6000 Hz. Due to the
relatively low sensitivity Of the human ear to low frequencies, the “bass drum isn’t
perceived as very loud in spite of its tremendous level which exceeds that of the
entire orchestra. For a close microphone position, the sound pressure level (SPL) is
in the neighborhood of 100 dB. For pop music and jazz, the resonance membrane
(head) opposite the playing one is usually removed and the tailing sound of the head
damped through the use of carpeting and similar devices. The kick drum is smaller
than the bass drum and therefore sounds quite different.
The snare drum, also called the side drum, has a comparatively narrow dygamic
range and only a moderately high level. It may be played with snare springs against
the resonant, opposite head. The frequency response range is again highly dependent
on the playing method and may increase with loudness to as much as 15,000 Hz.
The cymbals have very sharp resonances up to 5000 Fiz, size ‘dependent. But
even beyond that the frequency partialsare quite strong. For brushed cymbals, the
Percussion instruments and piano o/
inp a
LM TTT
OOOO EEL
drum I TE
|
salbey MN RIAL EEL ALOESAECL MEL EDEEE ELLE, CELE E ID
CVD AISI GI ILI LI LLL LL LL LILLE Lf

| |
| triangle
Xylophone
|
20 50 100 500 1000 5000 10000 20000 Hz
Hig = highest level range

VZZ = medium level range
(
_] = lowest level range
A. Frequency response ranges along with the tonal ranges (timpani, xylophone)
and areas of strong levels
timpani
bass drum
snare drum
cymbals
triangle
bells
xylophone
tom-tom
8 Ke 15 20ms so 40 So Gi) 70 80 90 100 dB
attack time sound level dynamic range
B. Attack duration and dynamic range
a piste
.=
open lid
Percussion instruments and piano 59
resonances shift to just below 15,000 Hz.

The tom-toms have a tuned pitch which is not quite as recognizable as with the
timpani. There are larger, free-standing tom-toms as well as smaller ones usually
mounted atop the bass or kick drum. The congas are tuned similarly to the tom-toms,
but they are not played with a stick but with the bare hand. The same is true of the
smaller bongos that are likewise tunable.
The xylophone characteristically has an extremely short and noisy attack pattern.
By contrast to the piano, the decay is on the order of the room reverberation at the
low frequency end and in the high register even shorter. This probably explains the
fact that, especially with the xylophone, the room tone with its information about the
room itself, is especially audible.
Bells belong in the same family as the xylophone, e.g., the idiophones. These
are mostly percussion instruments of vibrating plates or shells (bells, gong, tom-tom,
cymbals, etc.), rods (xylophone, vibraphone, marimba, celesta, triangle, etc), tubes (bell
chimes), or instruments with other shapes (wood blocks, castanets, etc). These
instruments generally exhibit non-harmonic sound spectra. The bell exhibits a very
special phenomenon known as the strike sound, ieafat the moment the bell is struck,
one hears a tone other than the one during the decay phase.
On the piano the strings are likewise struck and the attack is therefore rather
short: 10-25 ms. The spectrum encompasses/a range from 3 kHz to beyond 10 kHz]
Formants are not sharply defined and mostly give a clue to the instrument’s maker.
Descriptive of the piano sound are the ngise partials which are caused by the striking
of the string but which die away quickly after that. These ngises are centered between
200 and 1000 Hz. In the lower register they are largely masked by the harmonics of
the tone but appear more prominently in the higher registers. The playing touch
hardly has any effect on the attack but more on the timbre of the decaying tone. A
special feature of the piano’s spectrum is thefspreading of its harmonic partialg}which
may not be found in either string or wind instruments. This effect becomes clearer,
the thicker and stiffer the string becomes in relationship to its length. Therefore, it
is most noticeable for higher tones and for smaller size instruments, thus demonstrating
the negative aspect of such small size instruments. In the most unfavorable cases the
pitch sensation actually splits into a dual pitch perception. Singe the piano tone is
composed solely of attack and decay, it is the length of the decay which becomes-most
important. It is comparable to the reverberation time of a room (+p.36). The decay
time; defined analogously to reverberation time, decreases with increasing pitch but,
even within a tone itself, with increasing frequency of harmonics. Pianos displaying
a bright, transparent sound, show a relatively short decay time of 20 s in the lower
register as compared to the usual 30-40 s.
The radiation direction of a concert grand shows relatively wide angles both
towards the side and upwards (fig.C). [The radiating angle for high frequency com-
ponents is surprisingly narrow. JFor a closed top, the higher sound components, which
give the sound its presence are diverted towards the keyboard. As a whole, closing
of the piano’s top results in a duller sound with less presence but doesn’t make it
noticeably softer.
60 Sound sources
Speaking and singing voice
Speaking voice
Dynamic range and level: Human speech is rather soft when compared to the
sound levels of musical instruments. As a guide, the following averaged maximum
levels are valid for a microphone distance of 60 cm (24"); at twice this distance they
are about 4 dB lower, and at half the distance 4 dB higher:
Averaged max- Normal speech :

imum levels softer _ louder Feud SSC) geclaRe Cy liaualcs
Men 60 dB 65 dB 76 dB 16 dB
Women 58 dB , 63 dB 68 dB 10 dB
If the speech sounds half as loud, this means a level decrease of 6-7 dB. Mur-
..mured speech lies another 5 dB below the level of soft speaking; very loud speaking
. about 5 dB above loud speech. The dynamic range for extreme forms of speech lies
at about 25 dB for men and 20 dB for women. These values are valid for a so-called
microphone voice, i.e., for a style of speaking which does not accentuate individual
words, does not drop ends of sentences, and may be described as appropriate for an
acting voice. In recordings, this sort of voice leads to a lower average level and with
it, in practice, to a greater problem concerning the loudness ratio of speech to music
in level metering. ‘
Level structure: The level structure is definitely impulse-like, e.g., strong, high
level spikes caused by explosive sounds determine the highest level values. Short
pauses between sentences, phrases, words, syllables, and phonemes interrupt the level
pattern. This results in an average level which is far below the peak level, i.e., on
average by as much as 12 dB or, in other words, at 25% modulation. The average
level of popular music is generally assumed to be 6 dB below its peak content (at
50%) for serious music these values are 18 dB below or at 12%. Since the loudness
approximates the average level, speech is perceived to be significantly softer than music
when its peaks are at the same level. Therefore, it is impossible to achieve a speech/
music loudness balance from a comparison of the peak level meter readings. Rather,
the average levels derived from the peak levels must be used to achieve such balance.
An approximate equal loudness between serious music and announcements is then
likely if the music indicates peak and the announcements a level 6 dB lower. Natu-
rally such numbers can only serve as an approximate guide; manner of speech and
style of music play a very important part in this. *
Frequency range: Fig.A shows the average spectra for male and female speech.
The individual spectra are rather similar; however, with decreasing level, more and
more high frequency partials fall below the threshold of hearing. The radiation of
frequencies below 100 Hz (men) and 200 Hz (women) are largely independent of
speech loudness; they mostly are dependent on the distance. Playback levels which
greatly deviate from the loudness level at the microphone during recording produce
disagreeable changes in the low frequency components of the voice; at high levels as
boomy. It is the sibilants, particularly the S sounds with a range beyond 15,000 Hz,
which have the greatest response range.
Tonal structures: The vowels are the musical components of speech; they have
a melodic line spectrum. Various resonant-like peaks in the spectrum, the so-called
speech formants, differentiate the individual vowels (fig.B). A response range of
Speaking and singing voice 61
i 1 loud
dah
normal 4
soft
- Coo -
0
6.8.10" Beh 2658 10> 2s ANG S10 Hz 6:8.10%)20d 68.1082" «4 16:8107
men women
A. Average speaking voice spectra
SCH 4-6 kHz recording from in front

S SVP INppAn Sneterco: from the side
F 10-12 kHz ---- from above
a 5—7 kHz --—:— from in back
—30
We 2 4 6810° 2 4°6810°Hz2 125 250 500 1000 2000 40008000Hz
B. Speech sound formants C. Sound coloration for various

radiation directions
aly Wet
good
WSS SNES SESS

60 KSSSN3SSSSBSS SSS
|
marginal
SwNY
r feoPOO SSSe sand to
AOS, :
ag ones
SS es noise level
YesPESR oe 3%
0 10 20 30 dB Ae"200°“500 Iai 5000 10000 Hz
D. Word and sentence intelligibility with background
L
noise and restricted band width
62 Sound sources
soprano
LL a
Bea ae ee a ees alto
eae
ee ee ena (ONO!
yaaa
eee ne bass
(Pere ce | ee ee ees
30 40 50 60 70 80 90 100 dB
E. Speaking voice dynamic range and sound levels
Speaking voice:
men women, children
110 - 165 Hz 220 - 330 Hz
a
2 e © ef
Singing voice: = e = om
= =
| +4
bass 82 (73)
- 330 (392) Hz XKSKRKKROS mo xXY | |
baritone 98 (87) - 392 (494) Hz RRR
AXP AX
|
tenor 124 (110) - 494 (587) Hz PRR Lees
|
alto 175 (165) - 699 (880) Hz
mezzo-soprano 220 (196) - 880 (1047) Hz
soprano 262 (247) - 1047 (1319) Hz
;
eee
Ngacegeees
normal pitch range Ss
extreme pitch range ee
F. Pitch ranges of the human voice (voice placement)

Speaking and singing voice 63
3600 Hz i ired for their true transmission (the response of long distance phone
lines). uch as L MN NG R are comprised of a line and a continuous
noise spectrum. Explosive sounds (B P GK D T Qu) and fricative sounds (S SH CH
F X Z) are pure noise spectra partially characterized by formant structures. The
relatively high levels of S and SH above 10 kHz may lead to overload in sound storage
and transmission systems due to the pre-emphasis used in them. The low level of the
F sound may, by comparison, be easily overlooked during editing.
Radiating characteristic: Due to the directional character of the human voice,
its level is about 5 dB lower towards the side, and 10 dB lower at the back of the
head. For a microphone level identical to that used in frontal recording, this means
an increase in diffuse sound level of 5 dB, respectively 10 dB when recording towards
the side or back of the head. The character of the sound is modified significantly, i.e.,
frequencies above 1000 Hz are radiated to the side and the back at lower level (fig.C).
The greater reverberation portion and the decrease in higher frequency sound compo-
nents have an effect similar to a considerably greater microphone distance in a frontal
recording. In rooms with little diffuse energy, and especially for close placement of
the microphone, a position of the microphone to the side may help to avoid the
overload created by sibilants.
Intelligibility: Fig.D shows the word and sentence intelligibility as a function of
the weighted signal-to-noise ratio and the upper limit of the transmission range.
Practically 100% word intelligibility is assured for an upper frequency limit of 5 kHz
and a weighted signal-to-noise ratio of 30 dB.
Singing voice
The acoustic differences between the singing and speaking voices are not as
great as would appear from the aural perception. Since only sounds with an harmonic
structure possess pitch, these are particularly accentuated and expanded when singing.
While the speaking voice pitch changes gradually and often, the singing voice is tied
to certain pitch steps. The formant positions are matched to a certain extent to the
pitch of the fundamental that generally lends the vocal character a certain darkening
effect. The co-called singing formant between 2800 and 3000 Hz is of considerable
meaning for the timbre of the voice. This is generally connected with an overall
amplification of the higher frequency sound components which then give the human
voice its ability to be heard even over a loud orchestral background. This does not
take place in speech. A hallmark of the trained voice also is its vibrato and tremolo.
An especially pronounced accentuation and stretching of the vowels, singing formant,
vibrato and greater loudness and dynamic range define, purely acoustically, a clearly
well trained voice.
Dynamic range and level: Both dynamic range and maximum level of the singing
voice are surely a function of the particular singer and type of music. High female
(soprano) and male (tenor) voices attain the widest dynamic range. The peak level
of a soprano for a normal microphone distance may reach over 100 dB SPL (fig.E).
Vocalranges and character: The pitch range of a human singing voice determines
its vocal range (fig.F). Soprano, alto, tenor and bass are the main ranges. The
suitability of a voice or soloist for a specific role is described by its character (e.g.,
dramatic soprano, coloratura soprano, lyric alto, Heldentenor, young lover, basso
buffo).
7 Uige — —
Cray sent iss

aa = 4 Can as Lee oils
» PReebul oar
© « ic . ;
oe rk rac Vane > aon sas fF
p=) telexe om” ah ~ 2) hee’ mere earn
; ‘ i lieky ard at] a) ghthee a my
at Cw
Be 49 uli - laden Gere ra)
Latics Immpeuud aeft 94 9 7% ~ainets) 2-0 SF) Se vals edhe
A
: ) i Ae Ie sees ub aee: .
tei 2 Ovaet ly) a oS toon wedi og) alll ed anligenln eget a
ciate &ts eg eeenarae see
7. ee
“Ss —8 # AD ek
ics } = ay © nc Gre et> yAT me wrt @ =" oe
Ss wire =a ae wae siento:
7 : < ¢he SVT ie : a wi i esa quan bgart weal
aor 4
by Rete ape Tage aeons pee,

6 & —odeib>S > gee a
7 = )Hey
Seayye ¢ys “ao que nities DE
te wre nh & : 4 Pa erecting tr Jia ae arte <* S

Va a i qu —— Ae ." nent Sheu | a) » ties ry aKa =
DL eS ee eee ee Oe foe mos ak say Meares POUGE

tl. i, ay ky ae are i
“4 7 : 3 1
-
a * -» : ohh: 7A
a %e Heatin Pilcy be » gr att aoe ee vir tag
— pa {hie Sg, 4 gue gleam iveead deat
PO ie Sed a ce MGs st *ats ee
Maite ’ ie Se) eee 6 ce vir
. + _ rel
ed na 2oOoevGy 4 ge 4 . So?) Ae
ama \go ii a om eo ak
a ‘ ~ ra ~
reuU «é re = = we = oa
6) _ oe (5 a Ast % .
=e =
ues OL SA an Pie mh e-eWeeh

ne 4 Tae Seow! Aly DO a Say,’F
- - a
; La =|» .
; -
= |
: a ~<a
MICROPHONE RECORDING 65
Microphone directional characteristics

Microphone frequency response
Special purpose microphones
Spacial hearing
Stereophony
Stereo signal monitoring
MS and XY microphone techniques
Multi-microphone techniques
Time-of-arrival stereo technique (AB)
Mixed stereo techniques
Support microphones
Dummy-head binaural technique
The recording of string instruments
The recording of wind instruments
The recording of percussion instruments
The recording of guitars
The recording of keyboard instruments
The recording of speech
The recording of vocal soloists and chorus
Esthetic principles in musical recording
It is virtually y impossible
imp to p provide individual recipes
p for theft
positioning g of
microphones near musical instruments and voices. On the one handJthe position of
‘microphones is a function of the type of musi¢,)the sound of a particular ae
the acoustics of the recording space and the properties of the microphone being used.
On the other hand, there are very clear sound demands prior to a recording which
then define the resulting sound.
For serious music, the esthetics largely follow the natural sound, i.e., that sound
which one would experience in the concert hall. Since electronic reproduction does
not permit the same experience which natural listening imparts, one must shape the
sound during the recording through use of the specific means available to the ton-
meister. The results are, therefore, not identical but a comparable sound experience;
usually a sound with greater presence that does not lose the effect of spacial sensa-
tion. It is possible to attain a greater clarity of the composition’s structure through
a balancing of the instruments. Its spacial positioning is also determined by sound
zsthetic considerations. Popular music, in the main, has no natural sound. The
sound in the studio itself does not correspond to the intended final sound. The sound
of the recording is only created by means of the electronic equipment available to the
tonmeister. As a result one is free to choose entirely different microphone techniques,
especially a much closer spacing between microphones and the instruments. Jazz and
folk music have developed a recording technique which lies about half way between
that for serious and for popular music. a
——
66 Microphone recording
Microphone directional characteristics
Receptor and transducer principle

It is the receptor and transducing principle that determines the properties of a
microphone.
The receptor principle specifies to what sound field magnitude the microphone
reacts. Sound pressure receptors are always omni-directional microphones. Sound
pressure gradient (sound pressure difference) receptors and sound velocity receptors
are always directional microphones. Only through the combination of two directional
microphones is it possible for gradient or velocity receptors to work like omni-direc-
tional microphones.
The transducer principle indicates how the acoustical fluctuations are changed
into electrical oscillations. In a capacitor microphone a membrane that is excited by
the motion of the air particles vibrates as one plate of a capacitor. In a dynamic mi-
crophone a coil in contact with a membrane vibrates in a constant permanent mag-
net field or a suspended ribbon follows the motion of the air particles. [The transducer
principle influences the frequency response, the transient fidelity, the magnitude of the
voltage produced, etc., but has no )effect on the directional characteristic. ]
Directional characteristic definition

The directional characteristic of a microphone provides the directivity at various
frequencies in a polar coordinate diagram, i.e., how large the attenuation is in a
particular direction when compared to the 0° direction. This information gives the
frequency response relationship between the different directions and the 0° direction
(principal axis) but not the frequency response in the 0° direction (+p.70). This
response must be provided separately. The directional characteristic and the directivity
are valid for the direct sound, i.e., in the so-called free sound field. The directional
effect in the diffuse Sound field is given by the directivity index.fTt is the level relation-
ship of the direct sound energy from the 0° direction to the total diffuse sound energy.
‘The directional characteristics are only valid within the reverberation radius; beyon
it the directional effect decreases.
Directional microphones may work according to the pressure gradient, velocity
or interference principles (+p.74). In practice we mostly use pressure gradient recep-
tors.
Omni-directional characteristic of pressure receptors

The microphone membrane forms one side of an acoustically closed microphone
capsule and is set into motion by the sound pressure variations. It is equally sensitive
to sound arriving from every direction as long as the microphone is small compared
to the wave length of the arriving sound. The microphone has an omni-directional
characteristic. Above 5 kHz, the sound is no longer diffracted entirely around the
microphone capsule due to the shortness of the wave length, so that the omni-direc-
tional characteristic transitions via a cardioid to a narrow pattern (fig.A). The direc-
tional effect at high frequencies is reenforced as a result of pressure build-up for
frontally arriving sounds and cancellation at the membrane for sounds arriving from
the side (interference principle). This directional behavior occurs only in that part of
the frequency response range in which the diffuse sound portion is relatively minor,
so that the practical results for the recording are not of major importance./ Boundary
surface microphones (+p.103) have an effectively zero-dimension membrane for sound
waves and, therefore, a totally omni-directional sound reception over the entire fre-
quency response range]
Microphone directional characteristics 67
180° [a membrane 180°
NRE
omni- figure-8
(pressure receptor) (pressure gradient receptor)
180° omni
180° figure-8
+
N N either
hess EE
S000 or
cardioid by acoustic delay cardioid by addition of
(pressure gradient receptor) omni and figure-8
(pressure gradient receptor)
A. Receptor principles and their directional

characteristics (measured values for
various frequencies) 8 kHz
Microphone recordin
cardioid super-cardioid hyper-cardioid figure-8
sensitivity at:
oP =o1dB =)e@ls —12 dB -00 dB
180° -oo dB =4)/2 Cis —6 dB 0 dB
B. Comparison of cardioid, super-cardioid, hyper-cardioid, figure-8
Sih ie ¢
/ N t 1 |
ate } =A) /
Ss ea Nc J
cardioid + cardioid - only Cardioid -

cardioid = omni cardioid = figure-8 cardioid small cardioid =
hyper-cardioid
C. Obtaining the directional characteristic in a dual

membrane microphone
eines\
\S\\
\ i cardioid
omni a A, shot-gun
ce Wola
pipies
; 1 9 20) 11 222
| multiple of
ee source reverb radius
omni sof
D. Equal direct-to-diffuse sound ratio for unequal microphone

distances, valid inside the reverberation radius
Microphone directional characteristic 69
Figure-8 characteristic
In directional microphones with figure-8 (bi-directional) characteristics, the
directional characteristic results from the fact that sounds reaching the membrane
from the sides produce equal but opposite pressure on both sides of the membrane
and, therefore, prevent any membrane excursion. Sound waves arriving from in front
or behind the microphone cause membrane excursion (fig.A). The figure-8 charac-
teristic normally is available only in ribbon microphones or in those microphones with
switchable directional characteristic. It is usually obtained through the superimposi-
tion of two opposing cardioid patterns (see below). There are only a few figure-8
microphone models with a single membrane. It is a hallmark of the figure-8 pattern
that the directional characteristic is highly frequency independent. It discriminates
against sound from the side best of all. When compared to the omni-directional
pattern the figure-8, like the cardioid, may be positioned 1.7 times as far from the
sound source without increasing the diffuse sound level beyond that of the’ omni-
-directional pattern (fig.D). Signals impinging on the membrane’s back are amnti-
phase to those impinging on its front.
Cardioid characteristic
The cardioid characteristic may be produced by means of three acoustic prin-
ciples (fig.A): in the acoustic delay method (a method now largely being replaced by
the method next described) the sound reaching the microphone’s back must traverse
a longer path dimensioned in such a way that frontal and back sound arrive at the
membrane simultaneously causing no membrane excursion while sound coming from
the side, by comparison to the figure-8 pattern, does not reach both membrane sides
at the same time. Superimposition of omni-directional and figure-8 likewise produces
a cardioid. This is done either by connecting the output voltages of two closely spaced
capsules additively or by designing part of a membrane to act as a pressure gradient
transducer with figure-8 characteristic because it is also open towards thé back. [To
accomplish this, the capsule’s backplate is perforated, }
Hyper- and super-cardioid

These microphones’ directional behavior may be viewed as an unsymmetrical
figure-8 pattern (fig.B). The hyper-cardioid, when directed at the sound source,
absorbs the least diffuse-sound of any directional characteristic. Therefore its distance
from the sound source may be increased by 18% when compared to the cardioid or
figure-8. Compared to the omni it may be increased two fold (fig.D). The super-
cardioid has a slightly wider frontal lobe but has a higher front-to-back attenuation.
Of all the patterns it accepts the smallest amount of sound energy from the rear
half-room. An even tighter directional behavior than the hyper- or super-cardioid is
provided by the lobe of a shot-gun microphone (+p.77).
Switchable directional patterns

Capacitor microphones may use the following principles to switch their direc-
tional patterns (fig.C). In purely mechanically switchable elements, the microphone
capsule is either{closed at the back (oma) penat the back (figure-8), propen via
an acoustical delay network (cardioid).} y a single membrane is needed for this
principle. In the dual membrane unit, two cardioids oriented in opposite directions,
but having a common electrode, are added in phase (omni), added out of phase
(figure-8), or only one membrane is operational (cardioid). Two adjacent, or
back-to-back separated cardioid capsules may be superimposed as in the dual mem-
brane microphone. Therefore, the omni pattern of switchable pattern microphones
is not a pressure transducer omni-directional microphone with cardioid-like high
frequency directional response, but rather has the high frequency directional behavior
of a figure-8.
Microphone frequency response
The frequency-response or frequency curve of a microphone is one of its more

important quality parameters even though excellent frequency response is a require-
ment that alone does not define a good microphone. Microphone technical data for
studio microphones provide the frequency response and its tolerance range; it normally
provides for a +2 dB deviation from the design parameter curve. The frequency
response is provided for the microphone’s 0° axis, or, in other words, for the principal
axis ) sensitivity for reception of a plane wave. Therefore it applies(only to direct
sound)from in front and only if the microphone is set up within the reverberation
radius (~p.26), which for small rooms includes the sound source’s near field. ‘Tt doesn’t
apply to diffuse sound and sound from other directions.
Directional characteristic frequency response

The directional plots of microphones show the relative microphone levels for
several frequencies from all directions about the microphone referenced to the 0° axis.
It clearly shows that the curves for the various frequencies do not coincide. { In
practical terms this simply means that the microphone displays different frequency
response characteristics at directions other than on axis.} As a result sound coming
from the side and diffuse sound acquire a coloration when compared to the direct
sound. This is readily demonstrated by having an announcer turn away from the
microphone, thereby raising the diffuse sound portion. This is why studio micro-
phones, which almost without exception have a largely linear frequency response
given in the technical data only for the on-axis direction, may have vastly differing
sound characteristics.
Fig.A shows the 90° ahd 180° frequency response curves for capacitor studio
microphones. For cardioid pattern microphones there is a slight roll-off at high
frequencies at 90°, a certain high frequency boost at 180° which altogether leads to
a mild diffuse response high frequency rise. However, overall this microphone type
comes close to the ideal which is direct and diffuse sounds equality. Dynamic two-way
microphones (see next page) fulfill this requirement even better. The figure-8 pattern,
from among all available patterns, may be termed practically frequency independent.
For the omni-directional pressure transducer pattern, there is a significant high fre-
quency difference at 90° but a severe difference at 180° compared to the 0° direc-
tien. However, this doesn’t normally lead to a diffuse sound response high frequency
droop because it is kept linear by means of a rise in the on-axis high end response
in the direct (free) field. This microphone frequency response matching to the diffuse
field is proper becausefomni-directional microphones are often set up at a greater
distance from the sound’source} If the direct sound field response is kept linear then
the pressure transducer diffuse sound response exhibits a severe high frequency roll-
off. The omni-directional pattern of microphones with switchable directional patterns
is created in most microphones throughthe superimposition of two:cardioids and is,
therefore, also a pressure gradient transducer. The gradient omni microphone has the
same response at 180° that it has on axis. Its on-axis response, by contrast to pressure
transducer omni-directional microphones, is linear or only slightly high frequency
accentuated and, therefore, the diffuse sound has a commensurate high end roll-off
(fig.B).
Direct and diffuse field frequency response

The directional pattern’s frequency dependent characteristic provides for an
Microphone frequency response 71
0,1 Oey 1 sy alle) 246) kHz
pressure receptor
0,1 65 4 5 10 20 kHz ea!

dual-membrane
pressure gradient rceptor
IOF
—5 :
pressure
a6 microphone %
“x
x
AP ea ee aa eerie
Hy } pressure gradient fice
ms20a icrophone
ee ty =a
er Ore AZ
ee a ee 8 kHz
0,1 Dial 5 10 20 A ai ey,
A. Idealized frequency response B. Comparison of the omni-
at 0°, 90° and 180° directional character-
istic of pressure and
pressure gradient
receptors
8 —8
50 100 200 500 1 z 5 WO WS 50 100 200 500 1 2 SeeOlG
Hz kHz Hz kHz
pressure gradient receptor pressure receptor
C. Frequency response in the direct sound field (——)

and diffuse sound field (- -- -)
30
20 kHz dB
5) Silla) + ~
condenser 20 alee, Tanta |

microphone 45K |
dynamic 10 8
microphone 5 ail
0)
0.25 ms microphone distance
54cm: 510 20 50100 200 1000Hz
10,8 cm: 25 50 100 250500 1000 5000 Hz
DaAttack“and decay ‘char: 5,4 cm: 50100 200 50010002000 10000 Hz
acteristics of dynamic
and condenser microphones
for a Spark clscharge E. Low frequency rise in pressure
gradient microphones in the
near field
sum of the near

membrane excursion -— and far field
pressure difference microphone level pressure differential components
after frequency
response correction
bs
SSeS
-
aed «eee frequency eee ey
near field component membrane excursion
pressure difference microphone level
after frequency
response correction ——
frequency
frequency
far field component distance from the sound source
F. Near field effect as the sum of the near and far field components
and the frequency response correction in the microphone
Microphone frequency response 73
on-axis arriving sound which differs more or less from the sound which impinges on
the microphone from all sides simultaneously. Since the directionality of a microphone
is primarily effective only inside the sound source’s reverberation radius, the diffuse
sound frequency response is generally valid also for on-axis sound sources which lie
beyond the reverberation radius. {This directional pattern frequency dependence,
therefore, contributes heavily to the overall sound quality of a microphone.] Studio
quality pressure gradient transducers have a relatively flat diffuse sound response,
mostly with a slight high frequency rise. The direct and diffuse frequency responses
largely coincide (fig.C). Pressure transducers, on the other hand, always have a diffuse
field frequency response which deviates. greatly from its direct field response. The
diffuse response is flat while the direct field frequency response exhibits a high end
rise of about 6 dB between 7 and 12 kHz. At close range this causes a disagreeable
presence quality typical for this kind of microphone.
Dynamic and capacitor microphones

The frequency response curves of high quality dynamic microphones are not sig-
nificantly different from those of capacitor microphones. They are often used for
popular music performance because they have the advantage of providing, at their
price level, freedom from overload distortion even for the high sound pressure levels
found here and because they may be somewhat more rugged and require no external
powering. Differences appear in the attack and decay characteristics of these micro-
phones (fig.D). The capacitor microphone follows the sound field undulations clearly
better than does the dynamic microphone.
Proximity effect
All velocity and pressure gradient microphones, i.e., all directional microphones
exhibit a so-called proximity effect. This results in a rise at low frequencies for sound
sources close to the microphone, that increases with decreasing distance from the
sound source as well as decreasing frequency. /The boost sets in as soon as the
sound-to-microphone distance becomes smaller than the wave length.} This results
from the fact that the pressure difference which causes the membrane’s excursion
consists of two components: the pressure difference between the membrane’s front
and back decreases independently of distance with decreasing frequency, something
which is compensated through a commensurate bass amplification in the microphone
(distant field portion). Added to this in the near field is a frequency independent,
additional pressure difference, caused by the severe pressure reduction with increasing
distance. The fact that the frequency independent component also passes through the
microphone’s bass amplification causes the bass rise in the near field (near field
portion) (fig.F). Fig.E shows the bass rise for three sound-source-to-microphone
distances: 54 cm (21"), 10.8 cm (4.25") and 5.4 cm (2.13"). Figure-8 polar patterns
have a proximity effect 6 dB greater then that of the cardioid. The reason for this is
the fact that the cardioid pattern consists of equal parts of pressure gradient transducer
figure-8 (high proximity effect) and pressure transducer omni (no proximity effect).
The proximity effect is therefore cut in half (-6 dB). This effect no longer playS any
practical role for microphone distances beyond SO cm (20").
Microphones which compensate for this effect through a roll-off ai low frequen-
cies are called soloist, vocal or close talking microphones (+p.74). Universal micro-
phones often are equipped with a speech/music switch (lows rolled otf/linear). So-
called two-system microphones are dynamic pressure gradient receptors combining
two transducers: one for the low, and the other for the high frequencies. The low-fre-
quency transducer delay section has a longer path length which results in proximity
effect avoidance and reduction of the frequency dependent nature of the directional
characteristic.
Special purpose microphones
There are several special purpose microphones which serve to satisfy unusual
applications but which, because of their properties, may only be used optimally in the
prescribed applications.
Soloist microphones
Soloist microphones, vocalist microphones or close talking microphones are
directional microphones for distances up to a maximum of 30 cm (12"). Since these
microphones respond to the pressure gradient (+p.73), they are equipped for the
compensation of low frequencies with a fixed, a switchable, or a multi-position switch-
able low frequency roll-off (fig.A). When labeled ‘as the speech/music switch, the
position "speech" provides for the low frequency roll-off. A defined bass roll-off
provides an optimum linear response only for one microphone distance which may be
seen in fig.B. The roll-off refers to a frequency of 100 Hz. Microphones with a fixed
bass roll-off or only a single position switch are generally dimensioned for a 10 dB
roll-off at a distance of about 10 cm (4"). For closer distances the low frequency
components are amplified, while for greater distances they are weakened. This allows
the microphone distance to be used as a timbre creating device.
Soloist microphones reproduce distant sound sources with a considerable lack of
bass and may therefore find application only for vocalists or instruments at a very close
microphone distance. Besides the directional soloist microphones, one may also use
pressure microphones but usually solely for playback type recording. Because of
their acoustical make-up they exhibit no proximity effect and give off a linear frontal
response at any distance. Soloist microphones are usually hand held and are therefore
internally shock mounted.
Noise canceling microphones

Noise canceling microphones are used for recording or transmitting speech in a
very noisy environment. They utilize a strong bass roll-off from 1000 Hz on down to
suppress unwanted noise interference at a microphone distance of 2 to 4 cm (0.75 to
1.5"). Vocal soloist microphones do so starting only at 300 - 500 Hz (fig.C).
Lavaliere microphones
Lavaliere microphones are designed as clip-on microphones for speech use. The
speaker has it either hung around the neck by a ribbon, or wears it clipped to his
clothing or tie. The strange acousticalbehavior against the chest is compensated by
means of a special equalization curve designed into the microphone itself (fig.D), so
that a response similar to that for a microphone at a distance of 1 m (39") from the
mouth is simulated. Because of this special equalization, such a.microphene may not
be used for any other but the designed purpose. Nevertheless it is possible to use such
lavaliere microphones as contact microphones, for instance, to achieve special effects
from string instruments. Because of their high end boost, they produce a very dense
string sound advantageous for pop music.
Lavaliere microphones are normally pressure transducers with omnidirectional
pattern bécause théy need to be insensitive to sounds that might be caused by the mi-
Special purpose microphones 75
} |
|| 1| |
+44cho
| +
| || 1
50 100 1000 2000 5000 10000 20000Hz
A. Frequency response of a vocal microphone

with switchable bass attenuation
level attenuation
18 20dB at100Hz
50 orcm Microphone distance

for linear frequency
B. Optimum vocalist microphone distance response
10
ield response
dB
/1.8')
distant field response

(100 cm/3.3 ft)
20
. Frequency re-
sponse of a noise
compensating
microphone AS 6<810> 22 4 6 8 104 Hz
permissible
high frequency
droop
ie pressure gradient
receptor ~ desired curve for 27 cm |
. Frequency re- '
(11") average mouth distance
sponse of a
lavaliere 16° 2 5 Oman z:
microphone
Microphone recordin
SS tO KHZ
ee eer eee | KA 180°
E. Directional charac- SS SS SSS

teristic and acousti-
cal function of a
shot-gun cn aN
tube membrane
opening 0.5 mm ¢
(20 mils)
microphone
amplifier
amplifier
microphone
slit 0.1 mm (4 mils)
F. Boundary surface
ean ban LLL Eee ILL Ee
(PZM")
level
boundary frequency
frequency
boundary frequency required

boundary diameter
surface of
30 Hz 5m
50 Hz 3m
100 Hz 1.50 m
200 Hz 0.75 m
G. Frequency response
and boundary surface 200 Hz 0.30 m
diameter 1000 Hz
Special purpose microphones Wi
crophone rubbing against clothing. A newly developed pressure gradient cardioid

microphone allows a 3 dB higher sound reenforcement level than does the pressure
microphone.
Line or shot-gun microphones

Shot-gun microphones have a narrow-lobe directional characteristic which
concentrates the acceptance angle to a narrow angle, especially at higher frequencies
(fig-E). Great care must be exercised in their use with moving sound sources which
may include singers and instruments. At greater distances these types of microphones
are only then useful if sufficient direct sound arrives, i.e., with short reverberation times
and in larger rooms. Shot-gun mikes find application where one is forced to use a
greater microphone distance (film, TV, live opera etc.). They then allow individual
sound sources to be level balanced against other sound sources, and make the sound
source appear with greater presence. They operate as pressure gradient transducers
with a tube extension in front of the capsule which narrows the acceptance angle above
1000 Hz. Because of the fact that sounds which impinge on the slotted microphone
tube from the side are superimposed with different phase relationships on the sound
wave which travels down the tube, side originating sounds are partially canceled
through interference (fig-E). This is why such microphones are also called interference
receptors.
Measurement microphones
Measurement or test microphones are designed for their particular application
to have an optimum frequency response, dynamic range, frequency range and long term
stability. The frequency response is linear either on axis (0°) (free field microphone),
or in the diffuse field (diffuse field microphone). The directional characteristic for
the free field type is therefore of no interest, and does not need to be frequency
independent as is the case with studio microphones. Measurement microphones have
vastly differing response ranges which may extend to 100 kHz, but also only to 8 kHz.
Furthermore the response range may reach to the lowest frequencies, going towards
0 Hz that may cause infra-sound interference in recording. Generally speaking they
are not suitable for studio recording.
Boundary surface microphones

These microphones, which first appeared under the trade name PZM (Pressure
Zone Microphone™, Crown Corporation), utilize the special acousticalconditions which
exist only a few mm in front of extended reflecting surfaces. For small microphones
with their membranes in the reflecting plane there results a linear, direct/diffuse
sound, identical frequency response with a hemispherical, frequency independent
directional characteristic. The sound level is 6 dB higher than in the free field. For
one model of this type of microphone, the membrane is not co-planar with the
reflecting surface but rather faces it closely spaced (fig.F). The boundary surface
diameter must be at least half the wave length of the lowest frequency to be trans-
mitted. If this condition is not met, the response drops at 6 dB/octave (fig.G). For
certain applications this need not be a disadvantage. The mounting of this microphone
on a large surface avoids the usual comb filter effect(aused by floor reflectiong for
the normal microphone stand setup (+2 to +6 dB) and even for low stands (about
+1 dB). The reflecting properties of the boundary surface itself, however, do affect
the behavior of the boundary surface microphone.
Special applications are recording of acoustically well balanced ensembles in good
halls in time-of-arrival (AB) stereo and recording of moving sound sources with
changing direct/diffuse sound ratio (~p.100). [The spacial sensation of these micro-
phones is quite impressive’]
Spacial hearing
The ability to hear spatially results from the interaction of directional and
distance perception. For directional perception in the horizontal plane it is the
difference between the signals arriving at both ears that are determining, while
directional perception in the vertical plane depends on the sound coloration which is
a function of its vertical angle. For the judging of distance, it is the loudness level,
timbre and the ratio between direct and diffuse sound which are determining. The
experience and practice of the hearing mechanism has a significant influence on the
localization accuracy. For presentation of stereophonic sound images via loudspeakers
or an earphone pair, our hearing mechanism uses acoustic properties different from
those used in natural hearing.
Spacial perception of a sound source

If a sound source is located in the horizontal auditory plane off center, there
appear time-of-arrival differences up to a maximum of 0.6 ms (fig.A) as well as
frequency dependent level differences and, with these, timbre differences between the
two signals arriving at the two ears. For the speech spectrum the level differences
amount to about 7 dB (fig.B). The same amount applies to music as well, depend-
ing on the particular spectrum. In general both the time-of-arrival and level difference
act in concert to impart localization. For sine wave signals, the level difference is a
function of the frequency: below 500 Hz there are virtually no level differences, while
they increase with rising frequency above 500 Hz. The time-of-arrival difference of
sine wave signals is sensed as a phase difference between the ears. They impart a
unique location only up to 800 Hz; above that multiple locations. All in all, the
localization of sine wave signals is difficult and often leads to disorientation. The
localizing accuracy in the horizontal plane depends on the type of signal. Broad band
and impulse type signals are localized most accurately (about + 1°), narrow band signals
least accurately (about +5° to 10°). Narrow band signals may lead to multiple localiza-
tion if the signal is of short duration or if the head makes no direction finding move-
ment. Steady tones are mostly localized at low frequencies through time-of-arrival
differences and for higher frequencies predominantly through level differences. For
broad band’steady signals, it is primarily level differences which determine the sound
source direction but also the time displacement of the signal envelope at the ears.
If a signal moves upward in the vertical auditory plane, the level and time-of-arri-
val differences remain basically unchanged. The information about the rising sound
source is mainly taken from the change in timbre caused by the diffracting effect at
the head and ear lobes (pinnae). Sounds emanating from different directions obtain
direction specific boosts in certain frequency bands, i.e., the so-called direction deter-
mining bands (fig.C). Narrow band signals cannot have such complex spectral changes
and therefore localization of such signals becomes difficult if not impossible. The
localizing accuracy in the vertical plane is far inferior to that in the horizontal plane;
it depends clearly on the familiarity with the sound source and on the experience of
the hearing mechanism, Unknown speaking voices may be localized to + 15° to 20°,
known ones to + 10°. White noise may be localized as accurately as +4°. Localization
in the vertical plane is not possible for narrow band signals; the signal direction is
determined from the signal frequency.
Spacial hearing "2
ear signal arrival

time difference
Ms
0,6 =
OS aS) :
b
0,4 SS
0.3 4p = = i
a
0,2 —— ——=-
A. Time-of-arrival diff- 0,1 ia

erences between the
signals at the two
0 30° 60° go°
ears
a. calculated
b. by experimentation deflection of sound -
with headphones source from center
level and
level difference of the
ear signals for speech
erence — i
~
ees nae Ss.4
es ee es es CS Cee
B. Level differences 0 2 40) GO) GO WO WO TO CO ior

between the signals
at the two ears deflection of sound
for speech source from center
in front behind in front above behind

Vi ROSS 20d
=~CIR
C. Direction deter-
$f
_—_—_—_—————— $+ —__ + +
mining bands for 250 500 1000 2000 4000 8000 16000 Hz
sensation of the
vertical plane frequency ——_—
=
6
o
{S)
iS
§
6 4
xe)
2 7
D. Connection between £ 2
the actual and es- 8
timated sound source
distance (announcer
in minimally
reflecting room) 0 D) 4 6 8 m 10
actual distance ———»>
left
30° clicks ,
see
sound event 20° 7 Sine wave
direction ue tone (327 Hz)
2 /
right of left
earlier earlier
E. Delay difference be- ——
tween two loudspeakers ~ 1.0 0.5 ms 1,0
for stereo reproduction —40°
and the resulting deflec- arrival time
tion of the phantom . difference
sound source —20
= 307
right
left
30°
sound event °
direction 20
right
louder
F. Level difference between —30 —20 -10 1On =20F cOrds

two loudspeakers for —10° arrival time
stereo reproduction ifferen
and the resulting deflec- ’ difference
tion of the phantom —20
sound source }
Soule
right
Spacial hearin 81
Distant hearing
The ability to hear distant sounds is based even more on auditory experience than
is the hearing in the vertical plane. For distances between 3 and 15 m (10 and 49 ft)
out-of-doors, it is the decrease by 4 to 6 dB for every doubling of the distance which
provides the best clue to the sound source’s distance. For greater distances it is the
attenuation of the higher frequencies through air absorption which plays the most
important role and in the near field the linear distortion of the spectrum resulting from
diffraction around the head. Distance sensation is improved markedly in enclosed
spaces through evaluation of the ratio of direct to diffuse sound, permitting recog-
nition of even small distance differences. Altogether the recognition of the sound
source’s distance underlies very complex factors making it difficult to give generalized
rules. On the one hand, the perceived accuracy is quite poor, and, on the other,
eee distance of the sound source results in underestimation of the distance
ig.D).
Spacial hearing with loudspeaker reproduction

In stereophonic reproduction the spacial distribution of auditory stimuli is enabled
through a special phenomenon of sound perception which is of no significance for the
hearing of natural sound sources—the creation of so-called phantom sound sources.
A phantom sound source is created exactly in the center between two stereo loudspea-
kers when both speakers give off identical signals simultaneously. For increasing time
delay and/or level difference between the loudspeakers, the phantom sound source
moves away from the center until it appears to come to rest in one of the loud-
speakers. The louder or earlier arriving signal determines toward which loudspea-
ker the phantom sound source moves. The required delay time for a particular
angular deviation (fig-E) depends somewhat on the type of signal. A time-of-arrival
difference of only 1 ms results in the greatest possible deviation of 30° from the center
for the usual stereo loudspeaker reproduction setup in an equilateral triangle with the
listener. Level differences (fig.F) lead to more stable phantom sound sources. In
practice, level differences of 15 to 20 dB permit phantom sound sources to deviate so
far to the side that they become stable, real sound sources located at one of the
loudspeakers. It is a condition for the formation of phantom sound sources that the
listener be located on the exact center line between the speakers. Even for small
deviations in location the phantom sound source wanders towards the nearer loud-
speaker. The so-called stereo horizon within which true localization of the phantom
sound source is possible is only about 20 to 40 cm (8" - 16") wide for normal stereo
setups (+p.89).
The theory of summation localization attempts to explain the spacial percep-
tion of stereophonically reproduced sound sources. It assumes that for the percep-
tion of phantom sound sources in loudspeaker reproduction, summed signals result
from the superimposition of the sound fields of both loudspeakers, that are not further
differentiated by the hearing mechanism. They are not identical, but are equivalent
to those signals of a natural sound source located at the phantom sound position. By
contrast, the association model goes back to prior experiences of the brain. In two
recognition processes which run sequentially, we first determine the sound source
location and then the nature of the sound source. For instance, those sound colora-
tions which provide directional information for a raised sound source or that result
from the superimposition of time displaced stereo loudspeaker signals (comb filter
curve) are canceled. By contrast to the theory of summation localization, the associa-
tion model allows the unchallengeable explanation of spacial hearing.
Stereophony
Stereo methods
The stereo sound image may be reproduced either in a playback room—and it
is then described as room referenced stereophony — or it may be reproduced via head-
phones directly to the ear entrance, and it is then called head referenced stereophony.
(Note: The word monaural, often used erroneously in place of the correct term
monophonic, would properly be used to describe someone able to hear only through
one ear.) The stereophonic image has a fixed relationship to a space, demands of the
listener a position within the stereo area (+p.88), and requires him to face the base
line between the speakers. In binaural reproduction, the image follows the head
motion—something which is perceived as unnatural. Reproduction via loudspeakers
of binaurally recorded signals has become possible only recently through use of the
diffuse field equalized dummy-head microphone. While stereophonic signals are
obtained through the use of at least two microphones, according to different recording
and microphone methods, binaural recordings must use a dummy head, which, so to
speak, represents the listener in the recording room (~p.108)(fig.A).
Recording methods for room referenced stereophony

Stereophonic recordings are made according to two different methods (fig.A).
In the time-of-arrival (AB) stereophony, the time delay between the signals of the two
microphones of up to about 1 ms causes a similar time delay at the two loudspeakers
and, with that, leads to phantom sound sources which are located on the base line away
from the center during playback. Signals without delay cause the phantom sound
source to be centered (+p.78). The time-of-arrival differences at the microphones are
caused by differing path lengths from the sound source to the microphones. The stereo
sound image in time-of-arrival stereophony has an impressive spaciousness, however
individual instruments may appear to shift position suddenly when playing different
tones. The localization and directional resolution is not as good as with intensity
stereophony (+p.98).
In intensity stereophony the intensity or level differences determine the orienta-
tion of a sound source in the stereo sound image. There are three microphone
systems: the MS, the XY and the multi-microphone system (fig.A). They may be
combined. The level differences in the MS and XY methods are created at the
microphones through selection of certain directional characteristic combinations of two
microphones very closely spaced at a single location. This arrangement of a dual
microphone is called a coincident microphone. The coincident microphone, therefore,
supplies two microphone signals which are termed either X and Y or M and S depend-
ing on the technique employed (+p.90). The X and Y or M and S signals exhibit no
time delay, only level differences, depending on the impinging direction of the sound,
the directional pattern selected, and the microphone amplifier gain differences. The
XY technique directly supplies the L for the left and R for the right channel, while
the MS microphone technique supplies the L and R signals only after matrixing
(formation of the sum and difference). The two microphone arrangements are theore-
tically equivalent. However, in practice one finds considerable differences as a result
of the frequency response of the directional characteristics of the microphones and the
variations available in the remote controllability of patterns.
The sole dependence on level differences between the microphones in intensity
stereophony corresponds to the principles used in the further audio processing by
means of a mixing consol —amplification,
e mixing, influencing of the image location—
Stereophony 83
sound source
recording room
-microphones
dummy head
—mixing equipment
loudspeakers
playback room
listener
headphones
Microphone technique:
MS and XY micro- multi-microphone AB microphone dummy head micro-
phone technique | technique technique phone technique
Recording technique:
i
intensity stereo recording time-of-arrival

technique recording technique
Stereo technique:
room related stereophony head related
stereophony
A. Stereo, recording and microphone techniques (level differences

Ap,
time-of-arrival differences At)
ooo ooo cso, Conon)

ie} 10)
3 ES
oom OOO TTT)
—F feat ft
COUT rrr)
multi-microphone technique
eee
MS, XY and AB microphone techniques
B. Stereo reproduction of an extended sound source

Microphone technique spacial properties especially

of the sound suited for
) Yn
ae wa L & o
Ss6) 86) f=
ob
(2) =
=~ S£
© w is} ol @ £ SC
Co
6“ow
ao. 2D [e) (8) =
a aE = we re ® cf DO
SHB,rel oy eles
Ge. TS)Bese) lsfe)ay . eces
6 3 25
x Ne
ay o£& Oo Gal a a iro; AE
coincidence microphone
technique
multi microphone
technique
time-of-arrival
technique
support microphone
technique
dummy head
technique
C. Spacial properties of the sound picture and its preferred areas of application to
the various stereo techniques
-re applied ae chile as the stereo signals in

the contro! room, transmission, recording and playback, bui
not as the microphone signals.
signal) are the microphone signals in intensity stereo accord-

ing to the MS microphone technique; they may be converted
to LR signals.
X (left channel) and Y (right channel) are the microphone
signals in intensity stereo according to the XY microphone
technique; X and Y correspond directly to the L and R sig-
nals.
A (left channel) and B (right channel) are the microphone
signals in the time-of-arrival stereo technique; A and B cor-
respond directly to the L and R signals.
A stereo signal is deemed compatible, if a recording results
from its mono reproduction - M = (L + R) \/2 - which is not
noticeably worse than a mono recording made under com-
parable conditions.
The L and R signals are coherent if they stem from the same
sound source. Frequency independent level and arrival time
differences may appear. The coherence remains intact even
in the presence of frequency independent phase shifts of
180° between L and R.
The correlation of a stereo signal is the measure of the|Forsinewavesignals
congruency between L and R, independent of their levels.|r=cosg is only avail-
The correlation coefficient r may vary in the range between __| able statistically
-1 and +1. for music and speech
D.Stereo signal concepts

EE
Bie CCDNGU ita wae 85
namely through level alteration of the signals (quantitative means), Even in time-of-
arrival stereophony the imaging direction of single microphones and pairs of micro-
phones is accomplished on the console not through delay corrections but through
level differences only.
A further microphone technique used in intensity stereophony is the close-up or
multi-microphone technique (+p.94). With this technique a separate microphone is
set up in the near field of every sound source, such as individual instruments or
instrumental groups, which picks up as little sound from other sources or diffuse sound
as possible. As a result the level differences between the individual microphones
become rather large. But the possibilities are commensurately great of influencing the
sound of every individual sound source, of placing it in its proper spacial location and
adding appropriate reverberation.
In today’s recording practice, the enumerated stereo recording and microphone
techniques are used both singly and in combination. Additionally various mixed
recording techniques are used (+p.100), which combine intensity and AB stereo. The
choice of which system to use depends both on the type of recording, the acoustics of
the recording space and the subjective judgement of the tonmeister.
The binaural or dummy-head, head referenced stereophony combines time-of-
arrival and level differences for the horizontal, and timbre differences (frequency
response differences) for the vertical localization of the sound sources during playback,
matching the conditions for spacial perception (+p.108). It always requires playback
by means of headphones. The limitations connected with this technique are more
than compensated by the impressive realism and true-to-life simulation of the original
sound images. The recently developed dummy-head technique may also be played
back via loudspeakers, thereby yielding a normal stereo sound image. This technique
cannot be combined with other stereo techniques and places great demands on main-
tenance of a linear frequency response and excellent signal-to-noise ratio during
recording, transmission and playback.
The distance from the sound sources during playback is altered when compared
to the situation in natural listening due to the limited sound source to microphone
distance. In the multi-microphone technique, all the sound sources sound equally
distant even if spatially displaced in depth. In the XY, MS and AB microphone
technique, sounds coming from the side sound more distant than centered sound
sources, i.e., the sound image appears to curve away at the edges during playback.
This adverse effect may be compensated through use of support microphones. It is
beneficial to arrange the sound sources in an arc when using a single microphone.
However this arrangement is often not possible in practice (fig.B).
Fig.C attempts an overview of the advantages of the several techniques as well
as their special suitability for the various applications. This is done with all due
reservations that really should be enumerated.
Relationship between the L and R signals

It is descriptive of stereo signals that L and R signals generally contain different,
similar and identical signal portions. It is for this reason that aside from the concepts
for the individual channels(L/R, M/S, X/Y, A/B) we also have formed concepts for
the measure of the similarity of the L and R signals (compatibility, coherence, correla-
tion) (fig.D). The compatibility of a stereo signal, or in other words its suitability for
monophonic reproduction, was an important requirement of a stereo recording during
stereo’s introductory phase. Today one can assume that wherever an interest in critical
listening to a recording exists, we also find a suitable monitoring system. Altogether
non-compatible stereo signals only rarely result in a good stereo recording.
oe)6 Microphone recording
Stereo signal monitoring
Stereophonic imaging is monitored acoustically through listening and optically

by means of program level meters, stereo oscilloscope and the correlation coefficient
meter. The optical means require interpretation, they cannot replace the acoustical
monitors, however they make error diagnosis simpler. In situations of insufficient
loudspeaker monitoring quality, such as in remote trucks (OB vans), the optical
monitors are often indispensible. For pure time-of-arrival stereophony, the optical
indicators are not meaningful or are meaningful only in a limited way. The correlation
coefficient meter and the stereo scope show phase differences but not delay differen-
ces. Since phase differences change with pitch, these indications are worthless. The
following descriptions, therefore, refer exclusively to intensity stereophony. For mixed
recording techniques—ORTF stereo, OSS stereo etc. (+p.103)—the indication of the
correlation coefficient meter and stereo scope have only limited usefulness because
it is adversely affected by the delay differences of these techniques. Use of these
optical monitoring devices with dummy-head stereo has no value whatsoever.
Program level meters

The program level meters permit the observation of level differences between
L and R. The following approximation may be useful: 6-10 dB difference in level
means location half way between center and the side with the higher level; 15-20 dB
level difference means location entirely from the higher level speaker (fig.A). Center
sound sources, meaning those sound sources which are imaged in the center of the
stereo base line, appear 3 dB louder than images at the sides due to acoustical addition
in the playback room (fig.B). They must therefore be reduced in level by that amount
provided this is possible with the available microphone arrangement. With electrical
addition of the stereo signals, i.e., for monophonic reproduction of a stereo recording,
the centered sound sources even appear 6 dB louder if set to the same level as side
located sound sources. So even for stereophonically proper reduction of centered
sound sources by 3 dB, the mono mix still results in a 3 dB rise for centered sound
images. For pure XY, MS or AB techniques, the overemphasis of the center image
can only be compensated through a change in musician seating arrangement. Use of
a 90° filter for mono formation avoids the additional rise of the mid image. This sort
of mono combination is used only for mono broadcasting of stereo recordings but never
for broadcast receivers.
Correlation coefficient meters

The correlation coefficient meter shows the phase differences betweenL and R
as the cosine of the phase difference, therefore providing information about the
similarity or equivalence of the L and R signals. This is done over a wide dynamic
range (-30 to +20 dB) independently of level so that level differences between L and
R are not shown. Since this meter shows only the phase differences between L and
R, it is unsuited for use with time-of-arrival stereo techniques. Its indication permits
a conclusion about the compatibility of the resulting mono signal but only for intensity
stereophony (MS, XY, and multi-microphone techniques). An estimate of the base
width and of a possible hole in the center is also possible. The indication of the
correlation meter may be even more precise if simultaneously observing the stereo
scope. The indication may be interpreted as follows (fig.D):
Stereo signal monitoring 87
image location _
in the center
image location
half left §—=———
A. Level difference
and image location
on the stereo horizon
stereo reproduction mono reproduction
LandR Land R LandR only L

non-coherent non-coherent coherent
L=0d8 |R=0dB [M=0dB JM= +3 a8

dB]M=—-3
|
relative
| sound pressure
B. Reproduction of mid level (dB)
and side located forloud- |___ 7
sound sources in speaker — ——
stereo and mono reproduction
reproduction Cn Sano 43
ne
horizon width
hin ex]
spacing
listener
YX
‘stereo listening
| C. Stereo listening area A area
Microphone recordin
The L and R signals have

O identical levels but re-
O versed polarity. During
fe Pal stereo reproduction not
—
= @ localizable; during mono
a
fe) 7%) reproduction canceled out.
@ ror
L
rad o
Q
fe)
2
ror
a
5
42)
” A full level stereo
fe) signal with good
S
o
2® therefore
and
compatible horizon width.
”n
@
=a
r between 0 and +1
fel
Right channel missing
Ss (—-) or left
<q @ channel missing (- -- -).
@QD |
Both channels missing:
a
no picture at all.
Il [e)
fi
Stereo sound one channel
S) polarity reversed or
O with individual micro-
phone polarities reversed.
@ Sound is diffuse and not
7) mono compatible.
r between O and -1 unusable

signal
mono
faulty
signal
stereo
id
The L and R stereo signals
O are identical in level and
phase. This is a mono
O signal which will emanate
@ from the center during
stereo reproduction.
@
correlation coefficient stereo oscilloscope

meter
D. Stereo signals and their indication on the stereo oscilloscope and the correlation
coefficient meter
Stereo signal monitoring 89
Correlation +1: The L and R signals are identical, they show no phase coefficient
difference, may however have significantly differing levels. There
is a point source of sound within the sound picture that, even at
a location other than center, may lead to a correlation about 1.
The stereo signal is compatible without any question. Pure mono
recordings also have a correlation coefficient of +1.
Between This is a stereo signal with full stereo width which +0.3 and +0.7
+0.3 and +0.7: is compatible. A normal recording in intensity stereo has a reading
in this range.
0: The stereo signal contains no correlated signals but clearly defined
horizon extremes. There could be a hole in the middle. This
reading will also result if one channel is missing, for a mixture of
in and out of phase components, for two unrelated signals or no
signals at all.
Between This reading indicates a stereo recording with one channel or one
-0.7 and -0.3: microphone polarity reversed; the signal is not compatible.
-1: As with +1, however one channel polarity is reversed. A mono
combination will lead to almost total cancellation.
A negative correlation coefficient always points out a transmission error; either

a reversed polarity or in MS stereophony, an excessive S signal or, in the use of an
active pan-pot, a setting in the over-width area. In all of these cases, mono
compatibility is not possible at all or only in a limited way. For MS stereophony we
obtain negatively correlated signals from those directions in which the S signal is larger
than the M signal. This applies to most directional characteristic combinations for
signals outside the recording area; and also if the S signal is too large for the recording
situation in question.
Stereo scope
While the correlation coefficient meter specifically shows the compatibility of a
stereo signal as well as reversed polarity, the stereo scope permits judgement about
the stereo sound image itself with respect to the direction of the signal portions, base
width, reversed polarity, excessive width setting, level relationships, etc. It is especially
useful in poor loudspeaker monitoring conditions such as are found in remote trucks
(OB vans). The information provided by the stereo scope is clearer and easier to
interpret than that of the correlation coefficient meter (fig.D), especially when the
stereo signals are missing. Additionally, by contrast to the correlation meter, it
indicates the level differences between L and R and displays them on the screen as
a function of the image direction. Signals outside the +45° sector are in anti-phase.
Bent over spikes indicate limiting in only one channel. Phase differences between left
and right are shown as circles or ellipses on the screen.
Loudspeaker monitoring requirements

A reliable stereo sound image judgement requires a listening location within the
listening plane (fig.C). For the usual monitoring situation—loudspeakers 3 m (10 ft)
apart and distance from the monitoring position to each speaker likewise 3 m-—the
listening plane is not much wider than the listener’s head. Every deviation from the
center, therefore, leads to a deviation of the sound image to the closer loudspeaker.
The monitoring room must be acoustically so dead that neither reverberation (under
0.4 s) nor single reflections or shatter are audible.
MS and XY microphone techniques
M and §S, X and Y signals

The MS and the XY microphone techniques are two systems of intensity stereo
which deliver similar, comparable results. Both use a so-called coincident micro-
phone (fig.A). It consists of two adjacently located directional microphones (usually
capacitor) coaxially stacked one above the other with adjustable directional charac-
teristics. One microphone is fixed, the other usually rotatable. It is also possible to
arrange two independent microphones in this fashion. These two microphone methods
differ in the selected directional characteristics as well as the further signal processing.
If we assume mathematically precise, frequency independent directional characteristics
and transmission factors (sensitivity), we obtain identical equivalent resulting direc-
tional characteristic combinations for the MS and XY microphone techniques (fig.D).
Since this precondition can not be met in practice, and since furthermore in the MS,
but not in the XY system, all parameters are remote controllable, some differences
do result between the MS and XY microphone techniques in actual practice. The X
and Y signals directly yield the signals for the left and right channel, i.e., L and R.
They may be panned by means of the console’s two pan-pots. The position of these
pan-pots then provide the width limits of the stereo image during reproduction. The
terms X and Y are only applied to the microphone signals, L and R are the signals
of the mixing console. The M and S signals only supply L and R signals after the
forming of a sum and difference (matrixing) at the console (fig.B). The M signal
(mono, mid, sum, main signal) is directly usable as the mono signal. The S signal
(stereo, side, difference, directional signal) does not, in and of itself, represent any
usable signal; it solely contains the directional information. This MS technique is used
in FM broadcasting and in the stereo disk. In FM broadcasting the stereo signal is
broadcast as an MS signal converted by the receiver back into an LR signal. In TV
two channel audio transmission it is the R signal which is broadcast alongside the M
signal and the receiver is able to obtain the L signal from this (M-R). The pickup
stylus of the stereo disk player reproduces the M signal from its horizontal motion
(lateral) and the S signal from its hill-and-dale motion (vertical).
Setting up a coincident microphone pair

In the MS technique, the microphone which supplies the M signal may use any
available pattern. It is directed at the center of the sound to be recorded (fig.C). The
microphone for the S signal must be a figure-8 at all times. It is rotated 90° with
respect to the M signal and when looking towards the sound source, towards the left.
With the XY technique both microphones are always set to the same pattern—any
directional pattern other than the omni-directional pattern. They are rotated symmetri-
cally towards the left and right by an angle to be determined by the width of the sound
image to be recorded. For every recording it is the subtended angle which is import-
ant. The recording area is that subtended angle in the studio which the microphone
then reproduces across the entire stereo horizon on playback. It should, therefore,
be set in such a way as to fully encompass the sound image for extended width instru-
mental groups—no wider and no narrower. For sound sources of limited width, such
as one or two solo instruments, which generally should not be reproduced with full
stereo width, the recording area is selected to be wider than the width of the sound
source. Of course the microphone also accepts sound from outside the recording area
but those sound sources will either be reflected back at the edge of the stereo horizon
on the proper side or they will be imaged on the opposite side in anti-phase, i.e.,
MS and XY microphone techniques 91
instruments
A. Coincidence micro-
phone technique
B. The relationship M=(L+R):V2 t= (M+ S)\V2

between the
stereo signals S= Ul SRE R= (MieeSieV2
M, S,L and R
Description Assignment 0° direction MS microphone XY micro-

of system of stereo of the indivi- technique phone
signals dual system technique
SM 69, USM 69: S-signal: pattern: Y signal: pattern:

rotatable black dot; figure-8; orien- cardioid; figure-
system (a) CMTS: red dot; tation: 90° left. 8, perhaps super
es ee C 34: logo; cardoid; orienta-
wate Metee Sey.
system Seles Soten , C 422: “ red LED. tion:
ion: to right
i
(channel) Il Bxskss3
fixed Sees SM 69, USM 69: M signal: X-signal: pattern:

system Sek logo. pattern: cardioid, as in Y-signal;
M, X CTMS: red dot; omni or figure-8; orientation:
C 34, C 422: orientation: 0° to left, angle
logo. (center). as in Y-signal.
System
(channel) |
Orientation: Pattern control:

SM 69: remote control
sound source USM 69: local control
0° CMTS: local control
“90° left 90° right C 34: — remote control
Ate ee ———=——e— #8 C 422: ~remote control
C. Setting the directional characteristic and offset angle in coincidence Microphones
saa
Microphone recordin
stereo micro- stereo directional

phone technique recording area characteristic
of the resulting
A
mono signal
MS XY Mi = X= Y
Q a= 90"
P P :
[i aN
| l |
Ses ar
SL dee
2b Sree ere
» side reversed
side reversed
D. Equivalent directional characteristic combinations of the MS and XY techniques

with offset angle 9 and half the acceptance angle ¢ , as well as the commensurate
recording areas and the directional patterns which result in the mono formation from
the stereo signals
MS and XY microphone techniques 93
completely diffuse. It is, therefore, important that only diffuse sound reach the micro-
phone outside the recording area but never any direct sound from a sound source.
The recording area is often selected to be too large, and, as a result, the sound image
is reproduced too narrowly on the stereo horizon. Fig.D shows four different large
recording areas and the commensurate settings of the coincident microphones in MS
and XY technique. In the MS system, the M signal is usually larger but in no case
smaller than the S signal within the recording area. For setting 1b (fig.D) this is also
the case outside the recording area, but here leads to a side correct reflection of the
sound sources into the recording area, and thus leads to a doubling up of the stereo
horizon. The recording area is limited through the intersecting points of the M and
S directional characteristics (point P in fig.D). With the XY technique the directions
of least—not greatest—sensitivity determine the recording area. The X signal, with
the direction of its lowest sensitivity, determines the boundary of the right channel,
the Y signal, in a like manner, determines the left boundary of the recording area.
The acceptance or opening angle defines one half of the recording area. The so-called
offset angle for the two microphone systems indicates the number of degrees by which
the systems are to be rotated from the center axis towards the outside. In other words,
for the XY technique for a specific directional characteristic, it determines the record-
ing area (fig.D: 2a and 2b). In the MS system the relative microphone angle is always
90°, the offset angle of the M system always 0°. Here the recording area for a given
directional pattern may be influenced through variation of the S signal level. The
arrangement according to fig.D: 2d is also called Bliimlein stereophony after its British
inventor.
Diffuse sound
The equivalent mono signal (fig.D) provides information as to what degree the
diffuse sound, i.e., reverberation or perhaps audience noise, are recorded along with
the stereo signal. In the MS technique it is the M signal which is the mono signal.
In the XY technique the mono signal must be obtained from the sum of the X and
Y signals.
MS Microphone system practical advantages

In spite of the theoretical equality of the MS and XY techniques, practice has
shown certain advantages for the MS microphone technique:
= Controllability of the directional characteristic permits all adjustments of the
parameters to be remote controllable. For the XY system the relative angles
between the microphones must be adjusted at the microphone itself.
uw Inthe MS system, one of the microphones —the M microphone — always faces the
center of the sound image. In the XY system the microphones face the edges
of the sound source. For this reason the MS system more clearly accentuates the
brilliance of a recording. Only the MS technique allows the use of a pressure
transducer with its omni-directional characteristic as the M signal —something
which also raises the brilliance (+p.70). Bear in mind that dual-membrane
microphones are not pressure transducers even in their omni-directional setting.
w In many cases a conventional placement of the sound sources together with an
often desired, not excessively wide recording area, and good diffuse sound
attenuation (fig.D: 1c and 2c) cannot be satisfactorily realized using the XY
technique because of the fact that the required hyper-cardioid pattern is often
either quite frequency dependent or not even available.
Multi-microphone technique
Principle and application

The multi-microphone technique (fig.A) is one of the intensity stereo techniques
but differentiates itself from those using a coincident microphone. In the multi-micro-
phone technique the setup consists always of mono microphones placed close to the
sound sources. This may be very close, in the near field, or a bit further away as in
serious music. The level differences between L and R necessary for the directionality
are obtained through manipulation of pan-pots on the mixing console. The spacial
sensation missing because of the close microphone spacing is made up through the
addition of artificial reverberation and delay units. This technique permits wide sound
variations with regard to tonal balance, spacial sensation, frequency response, direc-
tionality and the application of special effects for each microphone. There is a gradual
transition from the multi-microphone technique to the technique using support
microphones (~p.104), something which makes a sharp delineation between the two
methods impossible. The multi-microphone technique is in wide use today; in the
popular music field it is virtually the only one used. It is also used in the serious music
field, along with the other techniques, especially in cooperative ventures with film, TV
and the phonograph record, although here with greater microphone distance than in
popular music. Only the multi-microphone technique, or perhaps to some extent a
technique using support microphones, is applicable to multi-track recording and there
offers the possibility of producing mix-downs according to very different asthetic sound
viewpoints.
A principal requirement of the multi-microphone technique is that each micro-
phone should only accept the sound source to which it is assigned, i.e., the attenuation
of other sound sources assigned to other microphones should be as great as possible.
The choice and setup location of suitable microphones results from this; also the rear-
rangement of the musicians’ seating to the extent permissible by performance practice.
In serious music there is little such possibility. But in popular music, jazz and other
studio productions, these setup freedoms are extensive.
Seating of the musicians

Soft sounding instruments require a relatively large amplification, loud ones a
relatively small one. If a soft instrument is located alongside a loud one, the loud
instrument will be transmitted through the microphone assigned to the soft one at too
high a level. In such a case the multi-microphone technique is no longer manage-
able. This is why instruments of equal loudness should be placed near one another.
In serious music, it is the optimum acoustic and eye contact, as well as the type of
music and traditional arrangement of instruments, which determines the microphone
placement and, therefore, the optimum application of multi-microphone technique
(fig.B) requires compromises. A seating arrangement in the form of an arc is often
beneficial as well as possible. For popular music it is possible to arrange the seating
of the musicians much more according to the requirements of the technical domain,
L.€., the points mentioned in fig.B may be much better fulfilled. The much used
playback method with its sequential recording of instruments may be ideally combined
with the multi-microphone technique.
Acoustical separation
The acoustical separation achieved in the studio determines the freedom of
choice in the mix-down. If the seating during the recording approximates the direc-
Multi-microphone technique 95
instruments 1|
mixing system
A. Multi-microphone L
technique L
- Equally loud instruments sit next to one another.
- Instrument spacing is as large as feasible; the acoustical and

optical contacts are to be respected in the studio, while additional
optical considerations for the audience apply to the stage.
- The instruments are to be arranged next to one another, not one

behind the other.
- lf instruments are arranged one behind the other, the loud ones
should be in front and the soft ones in back.
- lf the musicians face one another, the the good front-to-back

rejection of the cardioid may be used to advantage.
B. Preferred seating
of musicians
number of i}
microphones
40Q OF tracks Hee
2 = SiGls 2
4 = els B
8 -—9dB °
iG =A ols e 1
32 — Wo dB oO
3
0) 2 4 6 8 10 12 dB 14
sound level difference

between the two ——
sound sources
D. Level increase resulting from the addition of two levels
C. Influence of the number of microphones or the

number of tracks of a multi-track tape recorder
track tape machine on the program level of a
single channel (assumption: equal program
level of all individual channels)
ee
6 Microphone recordin
Pattern attenuation:
low- and mid- high

frequencies frequencies
cardioid ©
—_—s
45° ve linear frequency response
45> practically none minor (2-4 dB);
with large membrane
significant (7-9 dB).
90° significant (ca. 5dB) significant to major

(5-10 dB); with large
membrane severe
(to 15 dB)
135° severe (10-15 dB) very severe
180° (15-20 dB)
super-cardioid hyper-cardioid
t 180° very severe (15-20 dB) severe (10-15 dB)
yd 0° linear frequency response
45° minor (2-5 dB) clear

(3-8 dB)
90° clear (5-8 dB) severe
_(10-15 dB)
p30 135° severe (10-15 dB)
0° Et 180° rather severe (8-15 dB)

HSS?
shet-gun |,80° 0° frequency response slightly HF accentuated
45°
we 45° minor to clear
(2-10 dB)
90° severe
3 (10-20 dB) altogether
90 severely
135° very severe Asie
above 1 kHz (15-20 dB) ependent
~--------
below 1 kHz 192 : 180° (15-20 dB) rs
Q°
figure-8 45° OF linear frequency response
45° minor (to 5 dB)
go° very severe (20 dB)

_—_—_ 90°
135° minor (to 5 dB)
oe 180° linear frequency response

135°
-
rahe
E. Pattern attenuation of various microphone directional characteristics
ee
Multi-microphone technique 97
tional assignments in the final stereo image, then a lesser acoustical separation will
be sufficient. Any instrument in front of its microphone must be at least 6 dB louder
than any other instrument picked up through the same microphone (6 dB rule). This
is always guaranteed for instruments of equal loudness if the microphone distance to
its assigned instrument is not greater than 1/3 to 1/4 the distance to the next nearest
instrument. The microphone distance may not be reduced at will, since this causes
rather odd colorations. The use of separating screens (gobos) improves the acoustical
separation to about 10-15 dB, depending on the circumstance. In reverberant spaces
one may want to set up a gobo behind the musician to attenuate the diffuse sound.
Microphone type and setup

1. Basically only directional microphones are used. The following directional charac-
teristics are available (fig.E):
Cardioid: of all the principal directional polar patterns the cardioid has the highest
front-to-back rejection ratio, but the poorest front-to-side one.
Hyper-cardioid and super-cardioid: significantly lower front-to-back but far greater
front-to-side rejection than the cardioid.
Figure-8: of all the patterns the figure-8 has the best front-to-side rejection but it
is as sensitive in back as it is in front. It is therefore highly suitable for the pick-up
of two instruments which face one another which do not have to be individually
level controlled and which also may appear at the same location. It may also be
used for single instruments, provided no direct sound enters from the back. The
direct/diffuse (reverberant) sound ratio for frontal sound entry is identical to that
of the cardioid. It is often the sound of a microphone which plays a far more
important role in its selection than its particular directional characteristic. Because
of its largely frequency independent nature, the figure-8 pattern yields a less colored
diffuse sound than does the cardioid.
2. Since the microphones are positioned in the near field of the instruments or singers,
one must figure on very high sound pressure levels, especially for the brass and
percussion instruments. Modern professional capacitor microphones are well able
to handle such high sound pressures but are additionally equipped with an overload
protection switch (usually 10 dB) that may be actuated to reduce the input level
to, or the gain of, the microphone’s internal amplifier. For acoustical and other
reasons, dynamic microphones are often used in the popular music field.
3. Directional microphones normally display the proximity effectwith its low frequency
boost in the near field, something which is compensated for in the design of special
vocal microphones for microphone distances of about 10 cm (4")(+p.73). When
used at a greater distance this leads to a low frequency roll-off.
4. Of the special microphones the lavaliere is sometimes used for strings, plucked bass,
and acoustic guitar to obtain a sound with a more interesting presence. For iden-
tical suppression of the neighboring sound sources, the shot-gun microphone allows
a greater microphone distance than does the cardioid (fig-E). Boundary surface
microphones may also be used in the extremely near field, since they are pressure
transducers. Contact microphones are used more rarely (+p.74).
Individual microphone and overall levels

In the multi-microphone technique, every microphone or combined group of mi-
crophones is recorded on its own track normally at full level. For a mix-down during
the recording session, e.g., for stereo recordings or the later mix-down of the multi-
track tape recording, the levels of the individual microphones or tracks must be com-
bined with relatively low level, since, by comparison to other techniques, a great num-
ber of sources are added (fig.C and D).
Time-of-arrival stereo technique (AB)

In AB stereophony, the time-of-arrival differences between otherwise identical
signals from two microphones leads to time-of-arrival differences in the loudspeaker
signals (fig.A). This results in the formation of phantom sound sources between the
loudspeakers (+p.78,>p.82). To position the signal clearly at one of the loudspeakers
requires a time difference of 1 to 1.5 ms. The sound localization in AB stereophony
is considerably more dependent on the type of sound source than is the case for
intensity stereo. The ear is only able to differentiate delay in impulse type acoustical
structures but not in steady signals such as sine wave tones or the harmonics of
musical instruments. For these the ear registers the delay as phase differences which
are frequency dependent (they increase with increasing frequency for the same delay
difference), and which are unclear as to direction when they exceed 360°. As a result
the audible localization—especially for harmonic, steady sound sources (and this
includes most instruments, more or less)—is not precisely fixed and one obtains the
well known rapid shifting of direction for a change of pitch (fig.B). The presence of
all possible phase relationships between L and R is a hallmark of AB stereophony.
This results in good room reproduction in the recording. For a soloist who is not
located on the center line, for example, the phase relationships change constantly.
That is why the correlation coefficient meter and the stereo scope are not suitable
monitoring devices. The audible control alone is significant. This technique is
especially applicable if one wants to record an ensemble with impressive spacial
sensation but with limited localizability in an acoustically desirable space. Its applica-
tion is almost entirely in the recording of serious music.
For the mono formation from purely AB signals and for total compatibility in
time-of-arrival stereo, it is recommended that only the A or B signal be used by itself.
Since information about the recording technique used is normally not supplied with
a tape, this may seldom be applied in practice. The mono signal is then formed, as
in intensity stereophony, through the summation of L and R and then displays an
audible comb filter frequency response for non-centered sound sources (fig.C) when-
ever the inter-microphone distance is greater than 17.5 cm (7")(ear spacing).

The recording technique that uses time-of-arrival effects is called AB stereo-
phony. A pure AB stereophony exists only for microphones whose spacing is small
when compared to their distance to the sound source. A spacing of from 20 cm to 100
- 150 cm (8" to 40" - 60") is normal. For closer and extended width sound sources one
may use a closer spacing, for more distant and narrower ones, a wider spacing. The
use of omni-directional pressure microphones is sometimes preferred. But, in
reverberant rooms or with considerable audience noise one may wish to use cardioid
or figure-8 patterns.
If one wants to capture the room reverberation, one may wish to place am-
bience microphones. Then the AB time-of-arrival stereophony with a spacing of more
than 150 cm (60") provides an impressive image of the room acoustics. The great
microphone spacing assures that low pitched sound components are reproduced with
phase differences, e.g., diffusely. One may use cardioid microphones directed into
the hall from a location at the main microphones, or omni-directional microphones
at a greater distance.
Time-of-arrival stereo technique (AB) 99
path length or
arrival time
difference
A. Arrival time stereophony
sound
stimulus Z
direction :
B. Sound stimulus
710300 Hz
direction for
various frequency
sine wave tones
for a loudspeaker
arrangement of +30°
from the listener
frequency
3 \ only Aor B
—6 1 y
C. Mono signal fre- = y A+B
quency response of 42
5 1. partial can-
a pure arrival time 45
cellations
stereo recording
2. total can-
—18 cellations
Mixed stereo techniques

The intensity and AB stereophony recording techniques do not need to be used
separately, they may be combined (fig-A). Time-of-arrival and level differences
augment one another in their influence on the deviation of phantom sound sources
from the center. Fig.B shows the respective relationship of the equal effectiveness of
delay and of level differences when imaging a phantom sound source during play-
back. Mixed stereo recording techniques can serve to combine the advantages of
intensity stereo — precise localizability on the stereo horizon —and of AB stereophony —
good spacial sensation. The time-of-arrival differences assure the formation of
phantom sound sources even for diffuse sound low frequencies and, with it, the imaging
of the room on the stereo horizon. The level differences assure clearly perceivable
phantom sound sources in the higher frequency range where delay differences lead to
confused localization. Thus the two methods augment one another and, therefore, this
combination is often used in spite of the limited restriction on compatibility. In
practice one is only approximately able to judge which portion of the stereo image was
supplied by AB and which by the intensity method since too many factors exert their
influence (microphone type and directional characteristic, microphone orientation and
spacing, their distance from the sound source and location of the sound source). The
individual mixed recording techniques show significant differences in their intensity and
time-of-arrival portions.
Since the dummy-head method (+p.108) consists of a combination of delay and
intensity stereo, there exists a certain kinship between mixed recording methods and
the dummy-head method. Nevertheless there are certain differences. The intensity
differences in the dummy-head method are peculiarly frequency dependent and lead
to a multi-directional sound source imaging. In the usual intensity stereo method they
should be independent of frequency, something which is quite true in practice. The
dummy-head method is specifically intended for reproduction through headphones and
only they unfold its special advantages. The mixed methods are only meant for
reproduction through loudspeakers.
The same thing applies to the application of the mixed methods that applies to
XY, MS and AB techniques. They are most suitable for the recording of well balanced
ensembles in acoustically good spaces, i.e., especially for serious music and other music
using similar techniques. The simplicity of application of the mixed method is a great
advantage in recording. Such considerations as recording area, offset angle, choice
of directionality combinations (XY and MS method) and microphone spacing (AB
method) do not play an important role here. Only a suitable microphone position and
needed amplification must be determined. As a result, these methods are useful
especially when a suitable microphone setup must be found quickly in unfamiliar
surroundings.

In the mixed recording method there are a number of possible microphone
arrangements:
An often used one consists of two cardioid microphones and wide spacing the
way it is used in the AB method (+p.98). The more the microphones are toed out,
the greater will be the level difference (+p.90).
Two very special microphone methods in this arrangement worthy of mention are
the ORTF method and the OSS-method (Jecklin disk). Both methods are based on
Mixed stereo techniques 101
7 ee eee |
instruments
A. Mixed stereo
recording
techniques
dB
BO
soe :
fad IL IL right
acdpn : ian er
+4 half right
° 0°
=
g c middle
B. Equivalent values ae
§ | it 1s left
half
of level and ar- ‘=I
a : ; =
rival time diff- xo} 30°
erence between il leat r
Land Rwhenre- @ T left
produced on a
stereo horizon —2 -1 0 +1 +2ms
arrival time difference between L and R—*—
inter-aural
level
difference
C. Inter-aural level
differences as a
function of 20 \
frequency and the
subtended sound 15
source angle in the
horizontal plane
aks SS
(measured on a
sphere of 17.5 cm
(7") diameter)
0
0°
52
fi BS
ISINKS
fo] a 60 90
(=) from Cente,
D. ORTF microphone
technique using dual
microphone bracket
and twin capsule
microphone
30
cm
17.5cm
E. OSS microphone technique [~*~
instruments
F. Microphone technique
using 3 omni-direc-
tional characteristic
microphones
Mixed stereo techniques 103
ne delay caused by the 17 cm (7") distance between the ears associated with natural
earing.
The ORTF microphone method (fig.D) uses two cardioid microphones with a
spacing equal to the 17 cm (7") inter-aural distance. They are each toed out at an
angle of 55°. By contrast to the situation at the human head, the intensity differences
in this method are largely frequency independent. For a sound source location of
+45°, the level difference between the two microphones amounts to 6-8 dB. This
corresponds roughly to the values obtained at the human head (fig.C).
The OSS microphone method (Optimum Stereo System), also called Jecklin disk
(fig.E), is also based on the inter-aural distance but achieves it by means of a 30 cm
(12") diameter disk covered with absorptive material placed between the microphones.
This produces a frequency response change with rising frequency similar to that found
in natural hearing (fig.C). At low frequencies, up to about 500 Hz, and an angle of
+45° to +90°, one obtains level differences of about 5 dB which reduce between 1000
Hz and 1500 Hz due to the bright spot effect and which rises to between 6 and 10 dB
above 5000 Hz (+45°) and 8 to 12 dB (+90°). The bright spot-so called because of
the optical analogous effect —describes a phenomenon which shows a bright spot in the
center of the sound diffraction of a circular disk or sphere whose intensity and size
depends on the ratio between diameter and wave length. The OSS method uses two
pressure transducers (omni-directional characteristic).
The good recording attributes of pressure transducers that impart to a record-
ing good spaciousness and presence as well as natural sounding bass notes, has led
to different arrangements using several pressure microphones with greater spacing.
Three or four microphones may be arranged in a single line in front of a wide sound
source (fig.F). Other unusual arrangements are possible and are found in practical
applications. With these arrangements, frequency independent level differences are
obtained through highly different spacings to the individual microphones. The delay
times found here are considerably greater than in the pure time-of-arrival stereo
technique (+p.98).
Another mixed recording method uses arrangements of boundary surface micro-
phones (fig.F). Here, too, only pressure transducers are used. Annoying sound
colorations caused by minutely delayed reflections from an announcer’s table or from
the floor for suspended or stand-up microphones, are avoided due to the fact that the
microphone is built into a plate which is then placed on the table, floor or other
surface. The co-planar pressure transducer produces a hemispheric directional pattern
(-p.77). For stereo recording, two such microphones with a spacing of several meters
(yards) are placed on the floor, mounted against the side walls, or are suspended
attached to a large disk. Very good reproduction of room size, transparency and good
recording of moving sound sources (opera, drama, round table discussion) are advan-
tages of this technique. The excellent suitability for moving sound sources is based
on the absolute agreement between the direct and diffuse frequency responses of the
microphones. If an AB-like microphone spacing is used, then the boundary surface
method becomes a pure time-of-arrival method. By comparison to the AB method
using suspended microphones, it is the greater room tone presence and the comb filter
free frequency response that should be mentioned.
Support microphones
Improvement through the use of support microphones

Support microphones are used for the improvement of recordings, especially
under the following conditions:
Presence: Clarity, sharpness of contours, brilliance and closeness are the most import-
ant aspects of presence. Presence is achieved in recordings mainly through closely
placed microphones. In the multi-microphone method this is built right into the
method itself. Those methods which use centralized main microphone placements:
XY, MS, AB and the various mixed methods —achieve presence through the placing
of additional, mixed in support microphones. The sound-source-to-microphone spacing
here is somewhat greater than in the multi-microphone method, especially when one
support microphone —especially a stereo support microphone - must cover an entire
group of instruments. The closer spacing produces a greater diffuse sound attenuation
and good recording of a sound’s attack and noise properties which increases the
impression of closeness. The clarity is also increased due to the fact that the support
microphone signals arrive ahead of the main microphone signals, something, however,
which detracts from the spaciousness of the sound. If the spacial sensation and depth
perspective of the instruments are not to be adversely affected from the use of support
microphones, then the signals from these support microphones must be delayed by the
sound’s travel time to the main microphones. This might be possible for all support
microphones together, by applying an average delay time. The time delay is equal to
the time which the sound takes to travel from the support to the main microphones
(fig.A).
Tonal balance: Support microphones also provide the possibility to influence the
loudness balance of the individual instruments or instrument groups through a judi-
cious, musically knowledgeable operation of the mixing console. Musical structures—
for instance interesting sub voices—may be better accentuated through this method.
Serious music is normally composed for live performance before an audience. There-
fore, the instruments which are appropriate for the type of music in question are
properly loudness balanced against one another. In spite of this, there often appear
problems with tonal balance, when compared to the concert listener, caused by the
much closer microphone spacing which must be compensated through the use of
support microphones. For sound sources which are properly arranged as to depth
perspective, the closer instruments mask the more distant ones, something which does
not exist for a more distant live listener (fig.B). By contrast, in today’s popular music
it is standard procedure that the loudness balance is only created technically in the
control room. This is why only the multi-microphone method or a profusion of support
microphones may be used for such music.
Localizability: A further advantage of support microphones is that with their help
the localizability of individual sound sources may be improved. Even for those
techniques using more distant main microphones, it is usually prudent to increase
the location of the width extremes through the use of mono support microphones.
The assignment of the mono support microphones to any desired location on the
sound horizon, accomplished through the use of panorama potentiometers, is based
on the intensity stereo technique and, therefore, supplies the best localizing sharp-
ness. The depth perspective usually desirable in serious music recording, namely a
spacial depth of the recording area, is largely cancelled through the use of support
microphones. Localizability and presence may just not be united at a greater dis-
tance. There are two measures which can support an impression of greater distance
Support microphones 105
delay time to be set

]
oar 6 ee Oe, ta te ig 0 be on 6. | 8 re
OR eee te ON 27, 3.4 Ane RAO Ome Ot! 68 (De ORS, Cue Outt
path difference between main and support microphone
A. Mixed stereo recording techniques
paths to A and B
paths to A and B very different
nearly alike
B. Tonal balance change for small and large distances from an

extended size sound source
level difference
between microphones 3 6 9 20 dB
C. Waviness of the direct :

sound frequency re- waviness
sponse for a superimpo- TRO AVE i 10 > 2 dB
sition of microphone
signals
AS = 3.4m (11'2") 100 Hz 200 Hz 300 Hz

AS = 1.0 m (39") 340 Hz 680 Hz 1020 Hz
As = 34cm (13') 1000 Hz 2000 Hz 3000 Hz
+6 250
+3
0
D. Theoretical frequency response

behavior for two microphones at —12
aspacing _ s for identical iS
microphone levels (—) and
for a microphone level difference = [KG
of 10 dB (----)
strings
N XX I oo
Lo
orch.| |}{soloists}| orch.Il winds

\ Ny 7) U :
\ MW TT] x strings
\ \ iT] Je \ 1{ 7
\ AT " 7 N
\ \ I / ‘ rT //
\ VD / N " oa
\ ” / N " y
\ “uo N i 7
oo N u /
‘\ a /
7
recording area
-=--—————— imaging on the stereo horizon
E. Possible ways of applying support microphones (1.,2.) and several

main stereo microphones (3.,4.) to an extended size sound source
Support microphones 107
for the supported instruments: 1) the presence is reduced through a slight high
frequency roll-off, and 2) the localizability may be neutralized by mixing the mono
support microphone with like level but with a delay of 5 ms back into the stereo
summing bus.
Mono support microphones

Mono microphones are used when the sound sources have no particular width,
e.g., for one or two instruments belonging together, for vocal soloists and speakers.
Only mono support microphones may be used for relatively close microphone spacing.
To increase the desired selectivity of these microphones one may wish to use direc-
tional microphones (+p.66). Shot-gun microphones permit a 30 to 50% greater
microphone spacing for the same selectivity (+p.77).
Stereo support microphones

While mono support microphones are used for single instruments or vocal
soloists, the recording of wide-area ensembles (e.g., for individual sections of the
orchestra, for a chorus, and also for a solo keyboard instrument (piano, organ, harpsi-
chord)), in general requires stereo support microphones or several main equal level
stereo microphones. The poor rejection of back sound sources as well as the diffuse
reproduction of those sound sources lying outside the recording area is a basic problem
stemming from the use of stereo support microphones (~p.90). Therefore several
mono support microphones often may be used more advantageously. This is especially
true for instrumental sections seated one behind the other. The microphone accept-
ance area is matched to the width of the instrumental group, while the assignment of
direction, through the use of pan-pots and width controls on the console, must always
be seen from the position of the main microphone, and if numerous main microphones
are used, from the vantage point of the conductor (fig.E).
Sound coloration; the small room sound

The individual sound sources are recorded several times through the use of
support microphones or several main microphones which are at different distances
from the sound source (fig.C). Since these different delay time signals are mixed
together, the time-of-arrival differences of about 15 ms, equivalent to a 5 m (16 ft)
sound path difference, causes an uneven frequency response for the individual instru-
ments. This is the so-called comb filter curve formed from the addition and sub-trac-
tion of the two signals depending on their phase relationship (fig.D). A response in
which certain frequency components are missing altogether can only result from exactly
identical levels of both microphones. In actual practice, one of the microphones will
always have a somewhat higher level. This will reduce the ripple of the curve (fig.D).
Such a frequency response will also result if reflections from relatively close walls or
from the floor reach the microphone (~p.14). Since the maxima of this frequency
response are as regular as the harmonic structure of a tone, the resulting sound
coloration has a certain pitch characteristic, most noticeable for changes in distance
which produce changes of the pitch character, e.g., for moving sound sources. The
subjective sensation of this effect is often referred to as tubbiness.
A spacing between 5 and 10 m (16 and 33 ft) may transmit an impression of
room smallness. This results from the fact that reflections from a wall at a distance
of 2.5 to 5 m (8 to 16 ft) has a comparable effect. It is therefore important in the
application of support microphones that either the main or the support microphone
have a higher level by at least 6 dB so that the effect of the waviness of the frequen-
cy response or the small room sound remains negligible.
Dummy-head binaural technique
Principle
The dummy-head method was initially developed to permit judgment of room
sound. However, it has also found applications in radio drama and documentaries,
less in musical productions. The technique is normally intended solely for reproduction
through headphones. It provides an impressive all-around orientation with excellent
imaging of distances besides still some lingering weaknesses. It places great demands
on linearity and signal-to-noise ratio in transmission and storage media.
While experiments with dummy heads go back several decades, it was not until
1973 that the (now termed) old dummy-head method tried to obtain the same sort
of signals which impinge on the ear drums of an actual listener and to play these back
through headphones, once again to the ear drums, using a human head simulation with
two microphones installed in its ears. This required free field equalized headphones,
i.e., headphones which radiate the signal at the pinna (outer ear) with the same
frequency response that the signal would have if it came from in front (fig.B). How-
ever, such signals were not suitable for loudspeaker reproduction due to their frequen-
cy response. The new dummy-head method introduced in 1982 using a newly con-
ceived dummy-head differentiates itself from the old method in two important points:
(1) A simulation of the ear canal and ear drum is no longer used, while the outer ear
and head dimensions are matched to those of the average human as closely as possible.
(2) The mechanical dimensions of the coupling to the dummy-head’s ear produces a
linear frequency response for diffuse sound signals. The direct or free field signals,
however, have a response which is at all times a function of the direction (fig.C).
Reproduction
The headphones must have a response which will allow them to reproduce the
diffuse field without coloration. Headphones are normally free (direct) field equalized;
they have a flat response to signals from in front of the observer. The diffuse field
equalization of the dummy-head transmission is justifiable, on the one hand, because
a free field equalization is always possible solely for one direction, i.e., quite impos-
sible for stereo reproduction and, on the other, because the dummy-head is set up at
a greater distance from the sound source where the diffuse field portion (room tone,
reverberation) is considerable.
The dummy-head signals may now be reproduced directly via loudspeakers
because they are diffuse field linearized. As a stereo microphone, the dummy-head
supplies results which are comparable to a mixed recording technique (+p.100).
Application
The dummy-head may be used in the acoustical measurement field, for com-
parison of different room acoustics. This was its original application. In the recording
field, it has only found limited application. In view of the acoustic/documentary
character of its sound image, it is especially suited for feature and radio drama
productions. A combination with other recording techniques yields new dramatic
possibilities for headphone reproduction. Mono signals are in-head localized (thought
level), intensity stereo signals directly at the ear (whispering, conscience), and dummy-
head signals in the room (fig.D). In serious music production, this method is usually
used when the room plays a special role. Popular music productions can not use this
method effectively.
Dummy-head binaural technique 109
Hit | | | | |
time and le-
vel differences
for side location OR
ADORE
sound sources
loudspeaker SS ee Shea ene a

or a Sets ated ee E —
headphone Soe . —-
reproduction | = ee ae
20 50 100 200 500Hz1 2 5, 10) 20khiz
A. Dummy head stereophony; B. Frequency response of two free field
there are arrival time equalized headphones
and level differences
between L and R
sound impinging
direction
C. Frequency response of the

sound signal at the ear for
various subtended sound
source angles as well as for
diffuse sound. A headphone
must equalize a linear
response: for diffuse field
equalization per curve 1.;
per curve 2 = 0 UN (2)
200 400 600 Hz1 2 4 6 8 10 kHz
frequency ——s_
‘Tecording sound
technique imaging
@
-—mono in the head
intensit
D. Combination of venoue OS aS y atthe ead
recording techniques for
headphone reproduction = 8 \(—--~~—~~~—~—-~- piri head om
and their applicable aural in the room
sensation. el ee ee ee
The recording of string instruments
Directional characteristic, formants

The directional characteristic of string instruments (+p.44), their principal
attribute for the setting up of microphones, is characterized by less definition but
greater complexity compared to the wind instruments. The strings’ timbre changes
less with the microphone placement, therefore, the microphone setup is not as critical
as it is with wind instruments. Recording in the near field results in a curve com-
parable to a comb filter at higher frequencies due to irregular radiation (+p.44). With
the addition of artificial reverberation, this curve is also transferred to the frequency
response of the reverberation. A greater sound source to microphone spacing helps
to avoid this effect which leads to a sound shrillness. The main formant range (~p.44)
is important for reproducing the timbre of the particular instrument. The formant
range, however, is clearly defined only over a relatively small radiating angle. In the
violin the sound determining A-formant (about 1000 Hz) is principally radiated by the
top of the instrument’s sound box, so that the most favorable microphone location is
in this area as well. The microphone distance largely depends on the room acoustics,
the recording technique used and the esthetics of the recording (fig.A). The viola
displays similar characteristics to those of the violin. The violoncello radiates its low
frequencies principally in a direction at right angles to the front of the instrument.
It is directly in front of the instrument that it has its fullest sound. For a higher
positioned microphone, the sound becomes restricted since those frequencies which
impart that kind of sound quality (2000 - 5000 Hz) are radiated upwards (fig.A). As
a result, the celli often sound less full in the orchestra than do the violins. With the
contra-bass as well, it is best to aim the microphone at the body’s front, but not at the
f-holes. Typical for the sound of the double bass are the bow hair noise oscillations
which may reach frequencies up to a range of about 10,000 Hz. Their buzzing gives
the sound of the double bass a characteristic presence which may be heard at all times
over the sound of the orchestra. The otherwise poor localizability of the double bass
is improved significantly by this buzzing.
Extremely close recording

An extremely close placement of microphones to string instruments gives a very
present and dense sound image even in the presence of a great deal of reverbera-
tion. This sound is not a natural one and is, therefore, suited largely to the recording
of popular and similar types of music. In the presence of simultaneous sound reen-
forcement, the relatively low resulting microphone amplification improves the feedback
characteristics. Such a method may also be desirable on TV sets for optical reasons.
Lapel microphones are usually fastened to the tail piece or are aimed at the instrument
body. Such microphones may actually be clamped to the bridge of celli and double
basses due to these instruments’ larger size (fig.B). Lapel microphones display no low
frequency rise in the near field due to the fact that they are normally pressure transdu-
cers.
The fact that lapel microphones are often lavaliere microphones with their par-
ticular response curve is of no significance (+p.74). For extremely close microphone
placement one may also use general studio microphones on stands but this demands
cooperation from the musicians in view of the limits placed on their movements. For
celli and double basses, the microphones, wrapped in foam, may be clamped under
the bridge (fig.B). Vocalist and pressure microphones are preferred in this application.
Beyond that one may want to consider boundary surface microphones without their
mounting plate or even contact microphones.
The recording of string instruments 111
violin viola violoncello plucked bass
A. Microphone placement appropriate for string instruments
lapel microphone lapel microphone

on theontail piece
: on the bridge of F
lose-up microphone
of a violin or viola a cello or bass pore
under the bridgeP of
a contra bass
B. The use of lapel and close-up microphones for the recording of string
instruments
The recording of wind instruments
Directional characteristic, formants

When compared to the strings, wind instruments are very directionality depen-
dent in their radiation (~p.48,p.52). For the recording this means that the choice of
microphone position provides greater influence on the instruments’ timbre than does
the choice of microphone type. Often even a small change in the location of the
microphone or musician will result in significant changes in sound, and there are
microphone positions which will result in an unacceptable sound. In practice there
are two significant differences in the recording of woodwind and brass instruments.
Brass instruments, by contrast to the woodwinds, have a large bell which results
in a much greater directional concentration, even in the mid-range. On the other
hand, this larger bell provides for a greater acoustical coupling of the instrument to
the room resulting in greater sound energy radiation. Therefore, brass instruments
may be significantly louder than woodwinds. They radiate about 5 to 10 times the
sound energy meaning that they sound twice as loud as the woodwinds. The wood-
winds, unlike the brass, have finger holes along the length of the instrument, which
radiate significant portions of the sound. As a result, the directional characteristics
of woodwinds are more complex and are not rotationally symmetrical about the end
bell.
The bell direction of the instruments also is not uniform. They may point at
the floor (clarinet, oboe, soprano saxophone), more or less horizontally forward
(trumpet, trombone etc), horizontally towards the side (flute), upwards (bassoon, tuba)
or behind the player (french horn). For this reason there result different microphone
positions for the individual instruments as well.
It is rather strenuous to play a brass instrument, and, therefore, it is important
that one spare these musicians’ strength.
For the clarinet and oboe, sound components up to about 3000 Hz are predomi-
nantly radiated through the finger holes. Higher components radiate from the bell
(+p.48). This permits the selection of a brighter or darker sound through proper
orientation of the microphone (fig.A). In the area directly in front of the bell, the
sound is unnaturally tight and strident, and that is why this usually is not a microphone
position to be recommended. The microphone spacing is determined by the record-
ing technique employed. The closer the microphone, the greater the effect of even
slight motion of the instrument on the sound timbre. For this reason distances of less
than 50 cm (20") are recommended for use only with experienced studio musicians who
are able to keep their instruments in the predetermined position. If the microphone
is placed close to the mouthpiece, one obtains the disagreeable effect that each tone
seems to come from a different distance, depending on which keys happen to be open.
Normally the microphone is directed downward from above the instrument; in special
cases, €.g., to achieve better acoustical separation from neighboring instruments, it may
be directed at the instrument from the side, without any adverse effect on the sound
quality. The occasionally heard noisiness of the keys appears only with instruments
that are badly worn.
With the flute a microphone position over the keys yields satisfying results (fig.A).
The sound in front of the instrument’s end (it has no bell) is weak, noisy and tight.
Such a position is not recommended in practice, however, for pop and jazz, where
breathiness is often included in the musical intent, a recording made directly at the
mouthpiece may be desirable.
The recording of wind instruments iS
= -
full sound
darker sound breathy sound
brighter sound
oboe flute
clarinet
brighter sound
darker sound
saxophone
A. Microphone placement for woodwind instruments
2 Asana |
full to
\e dull sound
\
\
; strident
sound
\. indirect sound /
\ (serious music)
/
/
/
7
direct sound Ane
pop music ull to
( ) dull sound
french horn trumpet
gon sound with

presence
N
N
\
\ round
sound
strident
sound
trombone tuba
B. Microphone placement for brass instruments

The recording of wind instruments HS
The saxophone, whose bent-back end bell faces the keys themselves, permits
the recording of both the sound partials radiated from the finger holes and the bell
simultaneously. For microphone positions to the side, the sound becomes warmer
and gentler. Positions extremely close to the end bell result in a dull and tubby
sound. For the soprano saxophone with its in-line bell, the same parameters apply
as for the clarinet.
The bassoon radiates high frequency sound partials diagonally forward and
upward, the low frequency sound components, as in all woodwinds, to the side. The
recommended microphone position is the same as that for the oboe and clarinet.

The french horn is the only orchestral instrument which radiates its sound behind
the player (+p.55). Therefore, the french horn sound in the orchestra is always indirect
and rather spacious. Since this position of the instrument has remained unchanged
since it entered the orchestra, composers have used the french horn in a way which
corresponds to its unique sound. It is used as a sound integrating instrument often
intended to sound as if coming from a great distance. That is why a microphone
position in front of the bell is not recommended in serious music or for chamber music
(fig.B). Sound reflecting surfaces behind the french horn lead to disturbing reflections.
For the multi-microphone recording of popular music the french horn must also be
recorded in the near field, e.g., the microphone must be placed in front of the bell.
For semi-classical music, the setup used for serious music might be applicable. Even
though the nonlinear distortion in magnetic recording has been reduced markedly, it
is nevertheless the french horn which produces the most audible distortion at full level.
Similarly sensitive are the recorder, trombone ensembles, and children’s chorus.
The trumpet and trombone may be treated similarly when it comes to record-
ing (fig.B). The tone is the brightest on axis but, by contrast to the woodwinds, sounds
fairly agreeable. As one deviates from on-axis, they become increasingly dull (fig.B).
Both instruments achieve the orchestra’s highest loudness levels, with the exception
of a few percussion instruments. Sound pressure levels in excess of 120 dB are
obtained at a distance of 50 cm (20") in front of the bell; at 20 cm (8") over 135 dB.
This places their sound pressure levels into a range that cannot be reproduced distor-
tion-free by any but the most modern professional capacitor microphones. Significant
here is the overload sound pressure given in the microphone’s technical data for 0.5%
total harmonic distortion.
The tuba, the bass instrument of the brass, radiates its sound upwards like the
bassoon, the bass instrument of the woodwinds. The aggressive attack which typifies
the tuba’s sound is best obtained from a microphone position above the instrument.
Moving more towards the side causes a more rounded sound while embouchure
sounds decrease (fig.B). The Sousaphone is a tuba which radiates its sound forward
and is, therefore, most often used in marching bands.
The other horns radiate their sound in various ways depending on their con-
struction. If they have a trumpet shape (Fligelhorn, alto horn) forward; in their
Waldhorn shape (alto horn) towards the back; and in their tuba form (alto, tenor,
baritone horns) diagonally upward.
The recording of percussion instruments
Serious music
Far into the 19th century music accorded only a minor role to the percussion
instruments (+p.56). Only two or three timpani are noted regularly. From their
musical function, they act as the bass fundamentals of the trumpets. It is only during
the 19th century that this connection was dissolved. In order to record the timpani
with tonal precision, it is recommended that a support microphone be set up in the
near field, even for spatially oriented techniques such as XY, MS and AB. For
judicious mixing this permits a high degree of sound presence with a slightly earlier
attack than the orchestra’s.
The triangle does not require its own microphone. It is supposed to provide a
general, diffuse brightness to the orchestra, comparable to the Zimbelstern of the
pipe organ. The various drums are only then given their own support microphone
if they have some specific function, something one finds in many works of the 20th
century. The same thing is true for the cymbals, which display very high frequency
sound components, and which, for a fortissimo, may raise the entire level of the
orchestra considerably. This also holds true for the tom-tom. The support micro-
phones for these instruments are generally at a height of 1.5 m (59"). At this distance
only cardioid or hyper-cardioid patterns are conceivable.
Semi-classical and popular music

One finds the most varied possibilities in the recording of percussion (drums).
The method to be used depends not only on the type of music, the playing technique,
and the acoustical and technical situation, but also on the sound to be achieved and,
therefore, on taste trends. The range of possibilities extends from extremely near field
microphones for every instrument to a stereo microphone in intensity stereo or two
individual microphones for the entire drum set.
The most usual contemporary method of recording drums is to use a single
microphone for every one or two instruments (+p.94) and two overheads in addition;
two microphones 50 cm (20") above the cymbals, or a stereo microphone above the
entire arrangement. The stereo location assignment of the individual microphones
or instruments corresponds approximately to the natural seating of the instruments.
This is needed in any case when using overhead microphones in order to avoid double
imaging on the stereo horizon. But aside from this, the setup of the percussion
instruments (fig.A) provides an advantageous imaging for stereo: a) low-frequency
instruments (bass or kick drum, tom-toms) in the center because they are difficult to
localize anyway and because they would otherwise waste groove space on phonograph
records; b) high pitched instruments (cymbals, hi-hat) at the sides of the stereo horizon
because they make the stereo width especially clear. In general, the microphone
distance is or may be a bit greater in pure studio recordings than on stage with
simultaneous sound reenforcement. As to microphone type, dynamic microphones
are often used for subjective reasons or because of their robust construction, their
virtual freedom from overload and the fact that they require no powering. Because
of the fact that modern day capacitor microphones are equally robust, that their
overload reserve is usually sufficient, and even the simplest of consoles now offers
phantom powering, it is only the subjective sound judgement which should determine
the type of microphone. Due to the high sound pressure levels encountered, it is
recommended that the overload protection switch be used.
The recording of percussion instruments 117
A. Microphone placement for percussion instruments
1 cymbals 4 bass drum

2 standing tom-toms 5 snare drum
3 hanging tom-toms 6 high hat
bass drum with tom-torn,

resonance head snare,
removed other drums
{ > front head
front head
B. Microphone placement for drums
i
Microphone recordin
maracas cabaza pandeira
tambourine
claves tubo
guiro wood block gong
C. Less frequently used percussion instruments |

The recording of percussion instruments 119
In the bass or kick drum, the drum head opposite the one being played is
generally removed, permitting the microphone to be placed about 10 to 15 cm (4"-6")
behind the playing head. This provides a dry sound and good separation from the
other microphones. A blanket or foam material, possibly weighted down with a rock,
is often pressed against the drum head to obtain an even drier sound. The recom-
mended microphone is a dynamic one which is not proximity effect compensated
(+p.73). The bass drum has by far the highest level in the spectrum of a recording
and, therefore, determines its maximum level. That means that the bass drum balance
can only be increased at the expense of other instruments or voices. A desirable
reduction of its level without influencing its sound quality may only be achieved by
filtering out its lowest sound components below about 100 Hz. An electrical suppres-
sion of the sound hang-over is possible through use of a noise gate or expander and
the furthering of a rapid, precise attack is obtainable with a low threshold set limiter.
The side or snare drum is picked up at a distance of 5 to 10 cm (2"-4") from above
(fig.B). Important here is the lowest possible cross talk from the cymbals. For the
tom-tom the microphone is suspended over the drum head just as in the snare drum.
One microphone is well able to handle two tom-toms. The resonance drum head may
also be removed from the suspended tom-toms. Then the microphone may be inserted
from the bottom into the inside of the tom-tom (fig.B). To effect an even greater
damping of the snare drum or tom-tom skin, one often attaches a felt or foam strip
to the edge of the skin with adhesive tape. The cymbals microphones are suspended
30-50 cm (12-20") above the cymbals while for the hi-hat a lesser distance is permis-
sible.
Additional instruments to those listed, which form the nucleus of the percussion
group, may be added (fig.C). Gongs are recorded from behind at a distance of 30-80
cm (12"-32"). For closely spaced microphones, each gong gets its own microphone.
Bongos and congas as well as the tumbas (similar to the conga) are usually found in
pairs. If they are important then it is best to use two microphones per pair to allow
a stereo width expansion of the sound source. The individual microphones of the
percussion group are usually corrected through the use of filters or equalizers. For
the bass drum it is the high frequencies, for the cymbals and hi-hat it is the low
frequencies that are filtered out. Microphone setup and console settings generally
take quite a bit of time and care.
Jazz and folk music

Jazz recordings may be made using the same multi-microphone technique used
in popular music. If a well balanced sound reenforcement system is needed as well,
then only this method is applicable. Often the techniques applied in serious music
recording are used. Then one finds methods using main and support microphones,
or individual microphones set up at a somewhat greater distance. The method to be
used largely depends on the musical style and must be selected in consultation with
the musicians. This is also true for other types of music which fall into the area
between jazz and pop.
For recording folk music etc., the same things hold true that have been stated
for jazz. A method is often employed which falls somewhere between multi-micro-
phone and support microphone techniques.
The recording of guitars
Classical and acoustic guitars

Since the classic guitar has a rather soft sound, the microphone spacing must
be relatively close, even in serious music. About 50-100 cm (20"-40") is usual, other-
wise the acoustical and electrical signal-to-noise level ratios are too small. The
microphone is preferably to be aimed at the area below the bridge (fig.A). This
generally applies also to the acoustic guitar, a generic expression for the various
forms of the guitar without electrical amplification as used in the popular music field.
There are problems caused by this instrument’s low output, especially if a sound
reenforcement system is in operation during the recording or if the guitarist sings and
plays at the same time, requiring careful balancing. In such a case a lapel microphone
mounted to the sound hole, may bring the desired results. In those guitars which also
have a built-in pick-up, this is often mixed in with the microphone. In such a case it
is the pick-up which predominantly supplies the mid frequencies while the microphone
adds the low and high frequency components, so that it would be feasible to reduce
the microphone’s mid-frequencies by means of an equalizer to help the feedback
problem in sound reenforcement.
The same technique that is valid for the guitar also holds true for the mandolin
and lute. To pick up the harp, the microphone should be directed at its sound box.
Harp pedal noises usually stem from worn felt padding around the pedal slots.
Electric guitar
Only the amplification and reproduction of the guitar via loudspeakers has
permitted the guitar to become one of the most important instruments of modern
popular music. For one thing the amplification offers an opportunity to influence the
sound in many different ways using effects units, usually foot pedal controlled (fig.C).
Naturally the pick-up and the guitar amplifier as well as the construction and materials
of the instrument and strings have a significant influence on its sound as well. Dyna-
mic pick-ups must remain well away from magnetic fields such as a power isolation
transformer that might be located under the musician’s chair. Since the guitar should
sound the way the guitar player adjusts it, a pick-up with a microphone in front of the
guitar amplifier is the preferred method (fig.B). The microphone is to be set up in
the near field directly in front of the guitar amp loudspeaker facing the center of the
cone. For non-coaxial, two-way systems, two microphones are needed. Ifa directional
microphone is used at this close a distance, a bass roll-off must be provided either
by using a vocal microphone or by means of filtering. Direct injection recording of
the electrical signal from the pick-up often causes problems with hum due to ground
loops. Safety problems may be solved through the use of a direct injection box or
specially designed microphone amplifiers without a ground connection to the mixing
console. Often the direct signal is mixed with that of a microphone.
The electric bass, like the electric guitar, may be recorded with a microphone.
If the electric bass does not appear musically up front, it may be fed directly to the
console from a pickup. Of course both methods may also be combined. This is often
the case for a microphone pick-up when problems arise with feedback.
The recording of guitars 121
A. Classical or acoustic B. Amplifier microphone placement

guitar recording
chorus: voice doubling; subjective intensity increase.

compression sustainer: compressor which lengthens a tone without decreasing
level.
distortion: non-linear distortion with adjustable
properties.
flanging: time shifted signal overlay; varying delay time; vibrato
effect.
noise gate: shut-off for modulation pauses.
overdrive: tube amplifier type distortion; e.g. increases with
increasing level.
phasing: similar to flanging: phase shifted signal overlay;
frequency response of comb filter curve; aiso time
variable.
spectrum: adjustable boost in a steplessly adjustable frequency
range.
touch wah: automatically scanned filter with every tone.
Wwah-wah: individually controlled scanned filter.
a
C. Guitar effects devices

22, Microphone recording
The recording of keyboard instruments
Piano
A piano is generally any stringed keyboard instrument with a hammer action.
The proper name is actually pianoforte. For recording, one should always use a
so-called grand piano (225 cm (7’4") or longer). For serious music recording, unless
it only functions as a percussion instrument in the orchestra, the instrument is best
recorded in stereo and the room tone is to be included in such a recording. This
means a microphone spaced not too closely, and it may be done in MS, XY, AB or
one of the other techniques. There is no single preferred method; however, the
time-of-arrival and the mixed techniques appear to provide the best results. Many
singers prefer a closed piano lid. In such cases a position to the right of the pianist
or above the music stand provides the best presence, although disturbed to some extent
by page turning noises. Basically a closed lid provides no advantages and must be
avoided. A proper tonal balance is always possible with an open lid as well and
complete removal of the lid is best of all. (The hinge pins come out of every grand
piano to make removal simple). A closed lid always makes the grand piano sound
muddy and dull, making it virtually impossible to use any satisfactory stereo recording
technique (fig.A). For popular music it is recommended to place one microphone
each above the low and the high register strings at a distance of about 10-20 cm
(4"-8") and to use a pan-pot to obtain the proper stereo width. For monophonic
recording, a suitable microphone position is over the high strings or over one of the
front holes in the frame. Even though the sounding board, as viewed from below, is
attached to the frame, a microphone underneath the piano yields a dull and tubby
sound. The sounding board of a piano is quite thick (approx. 12 mm (0.5") and,
therefore, does not vibrate very much while the strings give off their high frequency
harmonics directly. For the upright piano, seldom used in recording, the lid must be
opened for popular music and the individual microphones placed as shown (fig.B).
One unsolvable problem found in recording a piano and a single singing voice
or instrument is the dual-roominess of such recordings. It always happens when two
sound sources are of unequal loudness and are brought to equal loudness at the
console. The louder sound source is then still represented more loudly in the rever-
beration signal. Therefore, the louder instrument always sounds further away and in
a larger room, the softer one always nearer. The dual-roominess is also caused by the
fact that the piano tone’s decay has the same structure as the reverberation of the
room, both spectrally and in duration. Thus the instrument’s own reverberation masks
the room reverberation and the roominess is only marginally audible. On the other
hand the singing voice, for instance, can cause the roominess to be especially clearly
audible; individual, impulse-like sounds generate single reflections which permit the
roominess to come out clearly causing the singer to appear to be in a larger room
than the piano.
The excellence of the instrument condition is of paramount importance in
recording (tuning, voicing). The piano make and, with it, the sound timbre of the
instrument do play an important part in practice. Certain makes of piano are prefer-
red by certain pianists and for certain kinds of compositions.
Other keyboard instruments

The celesta (fig.C), a keyboard instrument with the action of a piano but with
metal rods instead of strings and individual wooden resonators has often been required
in orchestras since the end of the 19th century. A microphone position at the back
The recording of keyboard instruments D23
serious music stereo close-up recording

lid open placement for upright pianos
serious music
lid closed
B. Upright piano recording
IM stereo
pop music
INS
stereo with multi-
microphone technique
s
sounding board
microphone (mono)
C. Close-up celesta recording

A. Grand piano recordings
microphone placement
stereo
placement
near fieldfone
microphones
various forms directional

e ; microphone at a
arpsichord E. Leslie speaker recording greater distanc
OW
a HW
PT PT
virginal BW
RP
spinet F. Organ layout

(according to the division principle)
HW Hauptwerk (great)
OW Oberwerk (swell)
BW Brustwerk (choir)
RP Ruckpositiv
D. Harpsichord PT Pedalturm (pedal tower)
The recording of keyboard instruments 125
of the instrument is preferred, in order to prevent the striking noise itself from
becoming overbearing in a recording.
The harpsichord is a keyboard instrument whose action plucks the strings. It
normally has the form of a grand piano, but smaller variations are also used. The
triangular spinet (more rarely five or six sided) and the rectangular virginal (fig.D)
are encountered more rarely. These instruments are now available in a more modern
construction with heavier materials and a sounding board open at the bottom as in
the piano, but of late there appears to be a trend back to traditional constructions
with a closed sounding box. The microphone setup is largely analogous to that used
for the piano (fig.D). The even loudness of the tones of these instruments is charac-
teristic for them. This may only be altered through changes in registration, articula-
tion and through the density of the musical composition as well as a comparably high
subjective loudness level for the same electrical output levels as other instruments.
The high loudness level is caused, in part, by the great sound density and a spectrum
extending up to the highest frequencies. Therefore, in practice the harpsichord is often
peaked at only 50% of full output level (-6 dB). By contrast, the clavichord is among
the softest instruments in music altogether. Here the recommendation is to place a
microphone very close over the tiny sounding board located at the right hand end of
the instrument.
Keyboards
The keyboard instruments in popular music, aside from the piano, include the
electronic (electronic organ, synthesizer, strings, etc.) or electro-mechanical instru-
ments (electronic pianos, Hammond organ). The sound is given off through loud-
speakers or is available for recording as an electrical signal. As in the electric guitar,
a microphone may be placed directly in front of the loudspeaker although direct
injection into the console without a microphone is preferred today (~p.120).
This is not possible when using a Leslie loudspeaker because the typical Leslie
sound is created through rotating elements in front of the loudspeakers. The rotation
produces a pitch vibrato due to the doppler effect which, if produced electronically,
does not yield the same sound. A slow rotation causes the cathedral effect, a fast one
the Leslie effect. At least two microphones must be set up, since the highs and lows
are radiated from two separate loudspeaker systems (fig.E). Wind screens are required
because of the air motion. However, it is better to place a highly directional micro-
phone at a greater distance, since otherwise the intended pitch vibrato is superim-
posed onto the loudness vibrato.
Pipe organ
The large pipe organ such as a church or concert organ is the largest instru-
ment considering outer dimensions and amount of material used. There are various
schemes for the arrangement of the organ pipes. Following the lead of baroque
instruments, groups of stops are combined into so-called chambers which are each
assigned to a certain manual (fig.F). Both for optical and acoustical reasons, the low
stops of the organ pedals are divided between two pedal towers, one left and one right,
in such a way that adjacent half notes are distributed to both towers. The result is that
the bass melody constantly jumps back and forth between the extremes of the organ
chambers. In organs of the 19th century, the visible pipes are arranged entirely for
optical/zsthetic considerations, often even using dummy pipes. The individual stops
are not combined into sound balanced chambers. Between these two versions any
conceivable combination of optical and acoustical pipe arrangement principles exist.
For large church organs the recording should impart a feeling of spaciousness, This
requires greater microphone spacing. There are no preferred recording techniques.
The recording of speech
Playback loudness and frequency response

The connection between playback loudness, natural loudness and coloration
exists in recording generally, but is particularly perceived in speech recording because
the human voice is among man’s best known sounds. The loudness of those voice
components below about 100 Hz for men, and 200 Hz for women is relatively indepen-
dent of the speaking volume (+p.60), and is therefore mainly determined by the
distance from the announcer. In every acoustic reproduction in which the playback
loudness level deviates from the natural loudness at the microphone location, there
results an unnatural sounding reproduction of the lows; at unnaturally high levels the
voice drones, because the lows are over-accentuated vis-a-vis the highs. For un-
naturally soft reproduction the sound becomes flat because the lows are missing
(fig.A).
Announcer recording
The sound pressure level for normal speech at a distance of about 60 cm (24")
from the announcer is about 60 dB which increases by about 4 dB to 64 dB when this
distance is cut in half. For loud speech, this level increases by about 6 dB. As a
result, in a properly treated broadcast studio, the unweighted signal-to-noise ratio to
the general studio and microphone noise amounts to about 50 dB. It is the self noise
level of the microphone which is the determining one. Since the studio and micro-
phone noises lie above the tape noise, short pauses in speech recordings must yield
mostly studio ambience. Therefore, it is recommended that a recording be made of
the studio atmosphere, consisting of studio, microphone and tape noise in case pauses
have to be added to the recording at a later time.
For relatively close microphone spacing—below 30 to 50 cm (12" to 20")—the
proximity effect (+p.70) and its noticeable low frequency boost provides a certain
boominess to the sound. For such cases there are microphones with switchable bass
roll-off, or microphones with a fixed low frequency droop, i.e., so-called vocal micro-
phones (+p.74). For the 60 cm (24") spacing particularly common in studios, this effect
does not play any significant role. Often the lavaliere microphone (+p.74) is used for
speech recording and is worn on the chest. In spite of its strange position, no adverse
response results. Much more annoying at close microphone distance is the popping
caused by the explosive sounds of the speaker. A so-called pop screen is the solution.
Disturbing sound colorations result when the microphone records reflections from the
desk or manuscript along with the direct sound; a choice of the proper physical
arrangement will help to avoid this (fig.B). Such sound colorations become disturb-
ing when the comb filter curve (+p.107) resulting from these reflections shifts as a
result of motion of the speech source. By contrast to dramatic studios, purely an-
nounce studios don’t have a minimum size. Proper selection of the speaker’s and
microphone’s location may provide room acoustics which will fulfill every acoustical
need. The reverberation time is generally 0.2 s to 0.3 s; first reflections should be
suppressed as much as possible.
Several demands are made on the announcer—some from the engineering side,
some from the problems of the medium. Strongly accentuated words at sentence
beginnings will result in a reduced average level, thus to a reduction in intelligibility,
and will have a negative effect on the loudness balance with other announcers or
music. Pauses between parts of the text, for instance, should be shorter than for a
lecture in front of an audience. Aside from this, recordings require greater discipline
The recording of speech 12 Z
amplification of softly spoken speech results in over-accentuation of the low frequencies
bass boost for...
natural
speech
loud
monitor level
bass boost
(= natural level)
frequency ——s_
A. Schematic presentation of the frequency response alteration

resulting from “unnaturally” loud monitoring
disadvantageous sound coloration no sound coloration

danger of microphone masking
no sound coloration
B. Microphone placement for

announcements or reading
in noisy surroundings
torus
for
mono
repro
duction
in very noisy surroundings
C. Microphone placement for an interview
= |
The recording of speech 129
with regard to noises such as script page turning. Because of the close microphone
spacing which acts as the surrogate for the listener when compared to the live situation,
care must be taken to suppress noises such as those caused by saliva, etc.
Interviews, eye witness accounts

A suitable interview microphone is principally selected for its directional charac-
teristic. An omni-directional microphone is suitable if one wishes to transmit the
acoustic atmosphere along with the speech. Omni microphones are also less suscept-
ible to wind, pop and finger noise than directional ones. The cardioid is well suited
for situations in which unwanted noises are to be suppressed and only the interviewer
or hissubject is to be recorded. The background noise determines how the micro-
phone is to be held (fig.C). A figure-8 theoretically attenuates background noise just
as well as a cardioid and is, therefore, well suited for the recording of two speakers,
but must then be held at mouth level. Due to the figure-8’s greater pattern integrity,
it will attenuate especially low frequency background sounds best of all the directional
patterns.
A wind and pop screen is always recommended. For microphones spaced closer
than 30 cm (12"), one should use a close talking microphone (see above). However
such a microphone falsifies the surrounding atmosphere. For very close talking, the
microphone’s membrane should not be addressed frontally, but rather at an angle to
avoid popping noise and overload. Directional microphones are rather sensitive to
mechanical noise interference. Therefore, it is important to avoid scraping or rubbing
sounds against the microphone or even the cable.
Round table discussions

The same factors that are true for single announcers, such as sound pressure
level, background noise, proximity effect and sound coloration apply to round table
discussions as well.
There are two possible microphone setups. In the first each of the speakers gets
a microphone according to the multi-microphone method (+p.94). The directionality
for stereo then must be assigned through the use of pan-pots on the console. This
method offers the possibility of opening a particular microphone only when it is
needed. This function may also be fulfilled by a so-called noise gate. To prevent the
appearance of an acoustic hole during pauses in the conversation, it is best to set up
a microphone solely for room-tone. For qualitative reasons this method is most
suitable to mono recordings. A better impression of the acoustical atmosphere in the
discussion room may be provided by two stereo microphones at a somewhat greater
spacing from the sound sources. The discussion partners are then best arranged on
an arc of 270°, while in XY technique, two cardioid patterns are cach rotated 45° from
the center line (fig.D). A setup with two microphones back-to-back is not recom-
mended.
A suitable setup for mono recordings is a stereo microphone with figure-8
patterns rotated at 90° to one another and combined through a 90° filter. This forms
the directional characteristic of a rotating figure-8 (torus or doughnut), i.e in the
horizontal plane equally sensitive all around but highly discriminating against diffuse
sound from above or below.
The recording of vocal soloists and chorus
Vocal soloists (popular music)

In this form of music the microphone-to-sound-source distance is basically very
small. One often uses a hand held microphone to give the singer freedom of move-
ment. The openings at the back of the microphone are not to be obstructed by the
hands because this transforms any directional characteristic into an omni-directional
pattern. If the vocalist also plays guitar or another instrument, then a microphone
stand or a lapel microphone connected with a wireless system are needed. The close
microphone spacing means high sound pressure levels, which, especially for explosive
sounds (b, p, d, t), leads to popping and for sibilant ones (s, sh, z) to a scratchy sound.
Both are caused by overload. The danger of such unwanted sounds largely may be
compensated if one does not address the membrane head-on but rather from the side
(fig.A). A pop or wind screen (or blimp) over the microphone also helps to reduce
such effects markedly. It is indispensible and is built right into those microphones
specifically intended for vocalist use. There are both dynamic and capacitor micro-
phones available as vocalist microphones that meet all of the requirements for a
close-talking microphone.
High quality dynamic microphones are often used in close talking applications.
They are rather overload proof on the one hand, on the other very rugged. While
directional microphones are normally used, one should also consider omni ones. The
advantages of using a pressure transducer are lower pop sensitivity due to a more
tightly stretched membrane, lower mechanical interference (e.g. from finger rubbing),
and no proximity effect. By contrast, the problem of sensitivity to sounds from all
around appears not to be too great a disadvantage, since high levels are produced
from the very close spacing. For soft voiced singers, however, the danger of feedback
from sound reenforcement systems may be too great. All directional microphones
display an increasing bass boost with decreasing distance, resulting from the proximity
effect (+p.73). For this reason, the microphones intended for vocalist use are either
equipped with a bass roll-off switch or with a fixed low end roll-off filter (+p.74).
For the recording of smaller vocal groups or background choruses, one can use
individual microphones under the same consideration as for individual soloists. The
advantage lies in the ease of balancing the individual voices. Since one usually works
with sheet music in studio recordings, it is recommended that music stands be used
to facilitate page turning. For an internally well balanced group, one microphone may
be used for two singers (fig.B).
Vocal soloists (serious music)

Hand microphones are never used for such singers. A microphone distance of
1 to 2 m (39" to 78") is preferred. Care must be taken in the microphone setup so
that no sheet music is placed in the sound path between the singer and the micro-
phone. Furthermore, attention will have to be paid to the concert situation in live
recording to make sure that the microphone does not cover the soloist’s face when
viewed from the audience. A microphone height at about the level of the sheet music
is preferred (fig.C), a height which avoids disturbing reflections from the sheet music
itself. When there are several soloists, they may sing into one microphone in pairs
(fig.D). Serious music soloists, especially sopranos, generate a considerable sound
pressure level, so that the dynamic range may get to be extensive -greater than for
most instruments (+p.63).
The recording of vocal soloists and chorus 131
2
eS
A. Hand held microphone advantageous not advantageous

for singers
B. Singing group |
C. Microphone for vocal soloists (serious music)

tenor bass
soprano alto
4 microphones
Lo) eo} ie} 1°}
3 microphones © le} io}
1 stereo microphone ()
2:
soprano tenor bass alto
4 microphones
(9) ie) (9) {eo}
3 microphones DO (9) {]
D. Placement of
choral sections 1 stereo microphone (6)
and microphones
more favorable placement
less favorable placement
E. Arrangement of
eee Se
choral singers
The recording of vocal soloists and chorus 133
Chorus
The arrangement of voices (+p.60) within a mixed chorus usually follows the
example shown in fig.D,1. It is advantageous with such a setup that those singers
from each section who stand near the center have good acoustic contact to all other
sections. In the performance this leads to a homogeneity of the choral sound. At
the same time, however, it also leads to a certain scrimming effect as for instance of
the counter-pontile structures. The arrangement according to fig.D,2 provides the
listener with the ability to differentiate acoustically between the individual voices.
This leads to increased transparency but, at the same time, makes it difficult for the
voices to sing together. This arrangement follows the same criteria of sound sym-
metry which also produced the German orchestra seating. An arrangement accord-
ing to the American seating is also possible: soprano - alto - tenor - bass. The singers
should be placed on risers to permit the free radiation of their voices towards the
microphone. Even a high microphone position cannot replace the riser method (fig.E).
There are various makeups for a chorus or choral composition:
Description Membership Number of Listing of voices Special groups;
of the chorus voices from top to bottom remarks
Mixed Women Usually 4, Soprano, alto chamber chorus
chorus (children) rarely 5, 6, _tenor,bass; if (small mixed choir)
and men or 8 more than 4 voices: double chorus,
soprano I and II (2 mixed choirs), a
cappella chorus (w/o
instrumental accom-
panimently)
Concert Usually amateur
chorus singers, secular and
church music with
orchestra;
Church choir Lay singers; church
music;
Opera professional singers;
chorus used in opera
: erformance
Women’s only female usually 3 soprano I, II
chorus voices and alto
Men’s cho- only male _ usually 4 tenor I, II and
Tus j bass I and II
Boys choir only boys 1 to 3 soprano I and II
voices; often and alto
as mixed
chorus with
men’s voices 4 to 8 as in mixed chorus
A stereo microphone is only then suitable, if the chorus does not stand behind
the orchestra. In such an arrangement, the orchestra would sound tco loud through
the choral microphone, and would prevent proper orchestra/chorus balance. Such
orchestra leakage would also tend to veil the orchestra sound. Three or four single
microphones would be preferable in such a case, and would allow the balancing of the
various voice groups.
Esthetic principles in musical recording
As with the zesthetics of all art forms, the esthetics of sound are forever in a state
of flux. This field, like all others, experiences short-lived trends or developments
connected with a particular individual. Nevertheless, it should be possible to formulate
some classical zsthetic principles in stereo recording. The treatment of the zsthetics
of sound gain in importance the more complex, the more spatially extensive a composi-
tion or performance is. In a recording of a singer with guitar it is relatively unim-
portant whether the singer is imaged to the right or to the left of the guitar, or whether
the guitar is a bit closer or more distant than the singer. In a recording of a large
work with vocal soloists, solo instruments, orchestra and chorus, on the other hand,
the zesthetics of the sound take on the dimensions of an important artistic question,
which, as with the performance of singers and musicians, is part and parcel of the
work’s interpretation. AEsthetic decisions about the sound usually start with relatively
small ensembles. They may not be made independently of the performance necessities
and, therefore, often take on the character of a compromise. Of course, it is a fact
that the traditional placement schemes were obviously all developed under considera-
tion of the sound zsthetic needs. A good example of this may be found in the
German orchestra seating (~p.34). It may well be that the current trend towards
greater precision in ensemble playing is often of greater importance, and this, in turn,
leads to the American seating.
In spite of all the changing aspects of sound zsthetic judgments, there are two
zesthetic principles applicable to stereo recording which are immutable: symmetry
and clarity. The transfer of these basic principles to the individual sound and space
dimensions of a stereo recording may be accomplished if the following guide lines
are followed.
Distribution of sound sources along the stereo horizon: The sound sources are sym-
metrically distributed along the stereo horizon, according to their pitch: high pitch
(left) - low pitch (center) - high pitch (right). This distribution is to be preferred to
the low —high—low arrangement because it is the higher frequency components which
more clearly define the flanks of the stereo image and because the low pitch instru-
ments do not let the problems of the phantom sound sources (~p.78) appear as
clearly. But the low pitched instruments also belong in the center functionally because
they form the common harmonic fundamental of all the instruments. In this particular
sense, it is the German seating which is preferred. There also is a great advantage
for the cutting of stereophonic records, since low frequencies waste the least space
when centered. Large works, for instance for vocal soloists, orchestra and chorus, have
several sound layers in the spacial depth, e.g., the vocal soloists in front slightly behind
them the orchestra and, behind it, the chorus. If the sound sources are well distributed
among the various distance layers, it clearly increases the clarity and transparency of
a stereo image. For this purpose there exist the following methods for assigning the
sound sources to two spacial layers, conforming to their pitch:
low — high —low high —low — high
high —low —high low — high-low
high — mid—low low — mid —high

low — mid — high high — mid —low
Esthetic principles in musical recording 135
For three perspective levels, the elements which are possible for two levels, may be
combined in many ways but at all times keeping the basic principles of symmetry and
clarity in mind. Here are two examples:
high —low — high high —- mid —low
low — high-low high-low —high
high —low — hi low — mid — high
A single solo instrument or singer is always placed at the center of the horizon.
The fact that in practice such sound-zsthetic considerations are often in juxtaposition
to performance based habits and demands in no way limits their validity. In niany
situations it should be easy at least to heighten the sound symmetry and clarity through
placement of the vocal soloists.
Width of the stereo horizon: The stereo width should not run counter to the room
perspective. Wide sound areas should be reproduced as wide as possible; narrow ones,
as for instance two instruments, are reproduced with decreasing width for increasing
distance of the sound source. A wide image horizon does not contradict the spatial
perspective of the listener only for closely spaced microphones. The room tone always
fills the entire horizon width independently of the reproduction width of the various
sound source.
Depth perspective: The smallest acoustically reproducible depth with loudspeaker

reproduction is the spacing of the loudspeakers; the largest possible acoustically
reproducible distance altogether lies between 10 and 20 m (33 ft. and 66 ft)(+p.78).
Since the distance perception ability is not as well developed as the directional per-
ception ability, it only is possible to differentiate between a very few distance planes
from smallest to largest reproducible distance. A well defined depth differentiation
in no way matches the experience of natural hearing, at least in the music field, but
rather primarily results as an acoustical perspective from the position of the main
microphone, and from the position of the conductor as well. The depth differentiation
offers the possibility of differentiating the sound space, a possibility which should be
utilized fully to compensate for the narrowing of the roominess which is unavoidable
in loudspeaker reproduction. Depth differentiation, at the same time, also means
differentiation of meaning. From our daily aural experience we learn that that which
is nearby is that which matters or even threatens; in short that which is more import-
ant than that which is distant. The depth differentiation realizable in a recording
may be that much more defined the larger the playing group and the larger the
perceived space. Acoustically it is not easy to generate depth perception (+p.104).
Semi-classics, popular music, folk music, jazz: The previously mentioned sound
zsthetic criteria are just as applicable to these types of music as they are to serious
music. The distribution of sound sources according to their pitch is perhaps even
more important here because the localizability on the stereo horizon, resulting from
the primarily multi-microphone and support microphone techniques, is even better.
The definition of the sound planes must be looked at somewhat differently. For
example, the rhythm group with drum set or rhythm and bass guitar or, perhaps even
the drum set alone, represents a sound plane. It is logical to place the bass drum in
the center, the cymbals and hi-hat are positioned at the extremes while the tom-toms
are arranged in ascending pitch order from left to right on the stereo horizon. The
spacial depth perspectives of this type of music does not have the same meaning as
it does for serious music if for no other reason but the fact that the ensembles are
much smaller.
136
Illustration sources
p.8,9: W.Kuhl: "Terminology of listening acoustics," produced by the Commit-

tee on Electro-Acoustics of the NTG (Association for Communications
Technology); Acustica, Vol.39 (1977) p.S7 (in German)
p.20 (fig.B): Acoustical information of the IRT (Broadcast Technical Institute) 226-1
(in German)
p.28 (fig.E), 34 (fig.E), 38 (fig.E), 46 (fig.C), 50 (fig.D), 54 (fig.D), 58
(fig.C): J. Meyer: Acoustics and Musical Performance Practice, (Frankfurt 1972,
Das Musikinstrument Publishing Co) (in English)
p.45 (fig.A), 46 (fig.D):
J. Meyer: The Physical Aspects of Violin Playing, (Sieburg 1978, Franz
Schmitt Publishing) (in German)
p.12 (fig.E):W. Kuhl: "The Combined Effect of Direct Sound, Early Reflections and
Reverberation on the Perception of Rooms and for Sound Recording,"
RTM (Broadcast Technical Information), Vol. 9 (1965), p.171 (in Ger-
man)
p.20 (fig.C):W. Furrer and A. Lauber: Room and Architectural Acoustics, Noise
Control, (Basel 1972, Birkhauser Publishing) (in German)
p.61 (fig-A):T. Tarnoczy: "The Average Energy Spectra of Speech;" Acustica, Vol. 24
(1971) p.65 (in German)
p.61 (fig.C):J.Meyer and A.H. Marshall: "Sound Radiation and Aural Impression for
a Singer;" report of the 13th Tonmeister Convention, Munich 1984 (in
German)
p72
(fig.C,D,E): G. Boré: Microphones; (G.Neumann/Gotham; 1973) (in English)
p.75 (fig.D):R. Plantz: "Electro-Acoustic Requirements for Lavaliere Microphones,"
RTM, Vol. 9 (1965), p.160. (in German)
INDEX 137
[A] Chamber music 48

AB stereophony 98 ensembles 31
Absorbers orchestra 32
effectiveness of 18 Chorus 133
high frequency 18 Clarinet 112
low frequency 21 Clarity 134
mid range 21 Classical guitar 120
porous 18 Clavichord 125
wedge-shaped 21 Coloration 70, 126
Absorption 6 Comb filter 17
coefficient 18 Concert
Absorptiveness 18 bands 52
Acceptance or opening angle 93 organ 125
Acoustic(al)(s) Congas 59, 119
aural 6 Contra-bass 110
delay method 69 Correlation coefficient
guitar 120 meters 86
panels 21 negative 89
room 6 Cymbals 56, 116, 119
separation 94 hi-hat 119
impedance 5
Acting voice 60 [D]
Ambience microphones 98 Decay, musical instrument 36, 73
Analyzer methods 5 Depth perspective 104, 135
Announce studio 126 Diffuse sound 6, 70, 93
Announcer 126 Direct and diffuse field 70
recording 126 Direct injection recording 120
Association model 81 Direct sound 6, 10, 13
Attack 36, 44, 48, 52, 73 Directed sound reflections 14
Attenuation 18 Directional characteristic 64, 66
Audience 21 cardioid 69, 97
figure-8 69, 73, 97
[B] hyper-cardioid 69, 97
Bands, big 44 narrow-lobe 77
Bassoon 115 . omni-directional 66
Bell direction 112 remote controllable 93
Bells 59 super-cardioid 69, 97
Binaural 85 switchable 69
Bongos 59, 119 Directional effect 66
Boundary Directivity 29, 66
surface diameter 77 index 66
microphones 103 statistical factor 29
Brass Distant hearing 81
bands 35, 48, 52 Drums 116
instruments 112 bass 56
instruments, historical 52 big 56
wind instruments 31, 115 bongos 59, 119
Brilliance 93 kick 119
Build-up 22 side 56
snare 56, 119
i? a | Tyialuornnminece 199
138 Index
[E] Intensity stereophony 82, 86

Dynamic (cont.) Interference receptors 77
pick-ups 120 Interviews 129
range 40, 47
level, 35, 47, 51) 55,63 [J]
program 40 Jazz 44, 48, 119, 135
Echo 6, 9, 17 Jecklin disk 103
flutter 17
Effects units 120 [K]
Electric Keyboard instruments 122, 125
bass 120
guitar 120 [L]
Enclosed spaces 81 Lapel microphones 110
Equivalent mono signal 93 Large room 17
Eye witness accounts 129 Lavaliere microphones 74
Laws of reflection 14
[F] Leslie loudspeaker 125
False localization 14 Level and
Figure-8 69, 73, 97 loudness 40
First reflections 6, 14 timbre dynamics 43
Flute 112 Level differences 100
Flutter echo 17 dynamics 43
Folk music 119, 135 Line sound source 10
Formants 39, 47, 51, 55 Listenability 9
range, main 110 Listening plane 89
positions 63 Localizability 104
singing 63 Localizing 13
French horn 115 accuracy 78
Frequency Loudness 40, 44
range 39, 44, 51, 55, 60 level 40
response 65 Low frequency
absorbers 21
[G] compensation of 74
Geometric propagation attenuation 10 Lute 120
Gobos 97
Gongs 119 [M]
M and S signals 90
(H] Mandolin 120
Harmonics 39 Measurement microphones 77
Harp 120 Mechanical noise interference 129
Harpsichord 125 Mechanically switchable elements 69
Head referenced stereophony 82, 85 Microphone
Headphones 108 acceptance area 107
Helmholtz resonators 21 ambience 98
Hi-hat 119 boundary 103
High frequency absorbers 18 capacitor 73
Hole in the center 86 clip-on 74
Horizontal auditory plane 78 close talking 73, 129
Humidity 10 coincident 82, 90
Hyper-cardioid 69, 97 conventional placement 93
distance 116
[I] dual membrane unit 69
Instrumental configurations 32 dynamic 73
Intelligibility 63 dynamic range 40
Index 159
Microphone (cont.) french horn 115

hand held 130 gongs 119
hyper-cardioid 69, 97 guitar 120
lapel 110 harp 120
lavaliere 74 harpsichord 125
location 13 keyboard 122
measurement 77 lute 120
mono support 107 mandolin 120
noise canceling 74 oboe 112
position 51 piano 56, 59, 122
pressure 74 pipe organ 125
shot-gun 77 saxophone 115
soloist 74 Strings 31
special purpose 65 timpani 56, 116
test “77 trombone 115
two-system 73 trumpet 115
type 116 tuba 115
voice 60 tumba 119
Microphone method viola 110
ORTF 103 violin 110
OSS 103 violoncello 110
Mid range absorbers 21 woodwinds 31, 112
Minimally reflecting rooms 21 xylophone 59
Mixed recording techniques 85
Mono formation 98 [N]
MS Narrow-lobe directional characteristic 77
stereophony 89 Near field 110
technique 90 Negative correlation coefficient 89
Multi-microphone technique 85 Noise
Music background 39
rock, jazz and popular 32 canceling microphones 74
semi-classical 48, 116, 135 components 47, 51, 55
serious 94, 108, 116 gate 129
Musical dynamics 43
Musical instrument 44, 48 [O]
acoustics 31 Oboe 112
decay 36, 73 Offset angle 93
dynamic range 31 Omni-directional characteristic 66
loudness 31 Opening angle 93
Musical instruments Orchestras 31
bassoon 115 symphony 32
bells 59 ORTF microphone method 103
bongos 59, 119 OSS microphone method 103
celesta 122 Over-width 89
clarinet 112
clavichord 125 [P]
congas 59, 119 Periodic sound events 5
contra-bass 110 Phantom sound sources 81
cymbals 56, 116, 119 Phase differences 86, 98
drums 56, 116, 119 Phon 40
electric Physical magnitudes 2
bass 120 Piano 56, 59, 122
guitar 120 Pipe organ 125
flute 112
140 Index
Pitch characteristic 107 Saxophone 115

Plane wave 2 Seating 21, 35, 94
Playback Separated cardioid capsules 69
dynamic range 40 Separating screens 97
loudness 126 Singing
Popping 126, 130 formant 63
Popular music 44, 94, 108, 116 voice 63
Porous Sones 40
absorbers 18 Sound
material 18 acoustics 4, 44, 48, 52, 56
Precedence effect 14, 104 analysis 5
Pressure clue 14
gradient transducers 73 coloration 70, 107, 126
microphones 74 continuous 17
receptors 66 diffraction around obstacles 13
transducers 73 diffuse 6, 70, 93
Program level meters 86 events
Proximity effect 73, 97, 126 non-periodic 5
periodic 5
[R] generation 44, 48, 52
Radiating characteristics 47, 51, 55, 63 impulses 17
Radio drama productions 108 presence 13
Real time frequency analyzer 5 pressure 2, 5
Receptor pressure gradient 5
principle 66 propagation 2
pressure 66 in air 2
Reduction 18 in solids 2
Resonant panel 21 over audience 13
Resonators, Helmholtz 21 reflections 6, 14
Reverberation 6, 17, 22 shadow 13
artificial 25 sources
build-up 6 center 86
decay 22 large area 10
duration 22 line 10
ideal 25 phantom 81
quasi stationery 22 point 10
radius 6, 26, 29 real 10
radius corrections 26 speed 5
ratio 26 velocity 5
time 6, 22 wave
Rise time behavior 13 properties 2
Rock, jazz and popular music 32 reflections 14
Room Spacial
impression 9 hearing 65
influences 6 impression 22
large 17 perception of sound source 78
minimally reflecting 21 Speaking voice 60
referenced stereophony 82 Spectral dynamics 43
size. 9; 17 Spherical wave 2
smallness 107 Standing waves 17
Room acoustics, fundamentals 6 Stereo
Round table discussions 129 methods 82
scope 89
[S] signal monitoring 65
Index 141
Stereo (cont.) Violoncello 110

sound image 89 Vocal
support microphones 107 groups 130
Stereophony 65 ranges 63
AB 98 soloists 130
head referenced 82, 85 Voice
intensity 82, 86 acting 60
MS 89 Voice (cont.)
room referenced 82 microphone 60
Strike sound 59 singing 63
String instruments 31 speaking 60
Summation localization, theory 81
Super-cardioid 69, 97 [W]
Surfaces with plantings 13 Wave
Symphony orchestra 32 form 39
lengths 5
[T] spherical 2
Technical dynamics 40 standing 17
Technique Wide-area ensembles 107
MS 90 Wind 10
mixed techniques 85 Wind and pop screen 129
multi-microphone 85 Woodwind instruments 31, 112
X Y technique 90
Temperature 10 [X]
Test chambers 21 X and Y signals 90
Timbre 13, 43, 44 X Y technique 90
Time analysis 36 Xylophone 59
Time-of-arrival
differences 98
stereo 82, 100
Timpani 56, 116
Tom-tom 59, 116, 119
Tonal
balance 104
range 44, 51, 55
structures 60
Transducer
pressure 73
pressure gradient 73
principle 66
Transparency 9
Tremolo 36
Triangle 116
Trombone 115
Trumpet 115
Tuba 115
Tubbiness 107
Tumba 119
[V]
Vertical auditory plane 78
Vibrato 36
Viola 110
Violin 110
Text set in Times Roman 12pt HP computer font type
reduced to 65% of original size;
Illustration text set in Helvetica 12pt reduced to 50%
using WordPerfect 5.0 text program;
Print-out on HP Laser Jet Series II printer.
Michael Dickreiter, Ph.D.
Born in 1942, Dickreiter studied 1962-1966 at the Tonmeister

Institute founded by Erich Thienhaus at Detmold, Germany as the
first institution to teach the Tonmeister profession. He subsequently
taught at the University of Valdivia, Chile. He received his Ph. D.
in Musicology from the University of Heidelberg with a dissertation
on the music theorist Johannes Keppler and minors in Physics and
Psychology. Since 1972 Dr. Dickreiter has been associated with the
teaching facility maintained by the German Broadcasting Systems
at Nuremberg as department head and author. Publications other
than this title include: Handbook of Studio Technology (in German),
Score Reading (in German) (also published in Japan), and many
others.
Stephen F. Temmer
Born in Vienna, Austria, where he was a member of the

Vienna Choir Boys, Mr. Temmer came to the United States 50 years
ago and continued to pursue intensive musical training with Moritz
Rosenthal, a pupil of Franz Liszt. Technical training followed both
at the high school and college levels, which led to an ABC career
as New York’s first tape recording engineer. There followed seven
years as co-owner and chief engineer of Gotham Recording Corp.
and in 1957 the take-over of the representation of NEUMANN,
EMT, Studer, Telefunken and other European professional lines in
the USA and Canada under the Gotham Audio Corporation name.
Mr. Temmer retired in 1985 to take up projects of interest such as
the translation of this book.
LLEGE
NEW BOOK
@

Tonmeister Technology Recording Environments Sound Sources and Microphone Techniques 096172000x 9780961720001 Compress

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Tonmeister Technology Recording Environments Sound Sources and Microphone Techniques 096172000x 9780961720001 Compress

Uploaded by

Copyright:

Available Formats

MICHAEL DICKREITER

Translated from the

Temmer Enterprises Inc. 1989

Title of the original German edition:

© 1984 S. Hirzel Verlag, Stuttgart, F.R. Germany

© 1989 Temmer Enterprises, Inc.

Printed in the United States of America

The areas of recording environment, Sound source, and microphone technique

Nuremberg, July 1983 Michael Dickreiter

What are Tonmeisters?

It is a precondition for the success of tonmeisters that they(have the ability

The reasons for this book

EL ROUEN TEV LERCOINS

Sound waves, fundamentals

Sound waves, fundamentals

Sound wave properties

Longitudinal wave (propagation in air

B. Sound waves in the

Plane wave in the distant sound field

frou Sede eae

MSC ud } = wave length (m)

sound pressure levei (dB)

For the plane wave: Q = air density (kg/m*)

Room acoustics, fundamentals

Influence of acoustical space on the sound event

Fundamentals of room acoustics

Fundamentals of aural acoustics

B. Direct sound and jdirect sound

short duration sound event

sound pressure sound pressure

larger distance from the sound source smaller distance from

CU) direct sound

— meowol total sound

-— Total sound level

Hil Direct sound

C. Distribution of direct and diffuse sound in an enclosed space

Application Concepts Secondary concepts Short definition

Speech and listenability . speech listenability acoustic suitability of a room

Music transparency . fange transparency clarity of a musical

Speech and room = . listener involve- sensation of the size and

D. Concepts of aural acoustics

Geometric propagation attenuation

Influences of temperature, humidity, and wind

1. point sound source (r = reference

dBy rel. sound level

dB ¥ rel. sound level

acoustic horizon sound rays

C. Diffracting effect level

D. Estimation of the size of the obstacle

E. Direct sound fre-

100 500 1000 Hz frequency

Sound diffraction around obstacles

So absorption for sound radiation over an audience seating area

The meaning of direct sound in recording

Sound wave reflections

Creation of sound reflections

Effect of sound reflections

True and false sound source localization

Increase in sound source loudness level

The sound is dis-

The sound is reflec-

If the sound arrives

D. The effect of mis-localization _ | yh j is

This loudness level increase is only of significance to a live listener, and is of no