acoustical principals of flute technique

8/12/2019 Acoustical Principals of Flute Technique

1/5

li Sfrumon HI st Volume XXVIII No 7 February 1974Sorne Acoustical Principies o Flute Technique

The flute is a very individualinstrument, its sound dependingin great measure upon the exactway in which the air jet from theplayers lips strikes the edge of theflute embouchure hale. Sorne players learn a good lip. techniquerelatively easily but for othersit is a constant struggle, and it issometimes difficult for a teacherto exphlin exactly what should bedone to improve the situation.Even for more advanced playersthere are constant problems instriving for flexibility. ftnd nuanceof timbre, and the variety of individual . lute tones amongst professional players attests to thediversity of possible solutions.

Much of the beauty and individuality of a musical sound depends upon the way in which it isattacked arid released and uponthe fluctuations associated withthevibrato. Indeed, when deprivedof their characteristic attack tran-sient, many instruments are almost unrecognizable - try 'playing a tape recording b a ~ k w a r d sDespite this, the steady tone quality is of immense importance, andit is impossible to produce a beau-tiful sound unless the basic steadytone is itself well developed andcontrolled.

During recent years out knowledge of the acoustical principiesgoverning tone production inwind instruments, and in flutesin particular, has increased verygreatly, although there still remainmany questions to be settled. Anexcellent account of our presentNeville H Fletcher is Professor ofPhysics at the University of New Englarzd in Armidale, Australia. He holdsa P h ~ D . degree frOm Harvard University and a D.Sc. from the Universityof Sydney both in the field of solidstate physics. He has studied flute inBoston arzd Sydney arzd played withsevera semi-professiorzal orche sttas.In the past few years he has combinedhis two interests through a study ofmusical acoustics which is being sup-ported by the Australian ResearchGrants Committee.

Neville H. Fletcherunderstanding of the acoustics ofthe flute has been given recentlyin these pages by John Coltmanland detailed references to theliterature can be found there, aswell as in a recent book by JohnBackus.2

The purpose of the present article is to take up severa aspectsof flute performance techniquewhich were ilot covered in Dr.Coltman's paper and to show how,by understanding the principiesunderlying sorne of the effects wetry to produce in performance, wemay perhaps make their attain-men easier. 1 would not claimthat this approach should supplanttraditional teaching methods, butitmay serve as a useful adjunct tothe more instinctive approachgenerally adopted.Basic Acoustical ProblemsThe simple motion of a flute. asa cylindrical tube, effectively openat each end, in which standingwaves can be excited by directingan air jet across a mouth-hole, sfamiliar enough not to require de-tailed discussion. The .possiblevibration modes of the pipe forman approximate harmonic serieswhich we can explore for ourselves by fingering low C C4)3 andthen, by appropriate blowing,sounding not only C4 but C5 , G5 ,CS, E6, G6 and perhaps even anote .clase to Bb6 . Which not_esounds depends upon lip positionand blowing pressure. A similarseries of notes can be _producedstarting from any fundamental,though the number ofnotes we canproducewithone fingering becomesfewer as we ascend the scale.l JohnW. eoltrnan, "Acoustics of the Flute,The lnstrumentalst XXVII/6 (January,1972), 36-40; XXVII/7 (February, 1972),37-43. .2. John Backus, The Acoustcai Foundationsof Music (New York: W.W .. Norton & eo .1969).3. I am using the U.S.A. Standard Notationin which the notes of the octave beginning at

~ 16 Hz are given superscript O, the loweste on a piano is e l etc.

Obviously, one of the most basicskills in flute technique is to learnto control just which note of thisseries will actually sound. However, we must recognize that whenever a note is played, all the higherovertones in the series for thatnote are excited to sorne extent.Since the relative intensities ofthese harmonics determine thetimbre of the sound, it is important to be able to control themwithin certain limits.

In addition to these tonal qualities, a player must be able to varythe sounding pitch ,slightly (tokeep in tune with other instru-ments) and must also have controlof the loudness of his sound. Tomake matters even more complicated. he should be able to control each of these variables independently of all the others, anearly impossible task.

For the expei'ienced player,all of these adjustments are in.stinctive and he thinks aboutthem little mdre than we think ofthe necessary muscular: acljustments we m.st make to keep ourbalance while running or walking. But for the beginner or evenfor the experienced player meetinga new technical or interpretivedifficultv. it is often a great helpto understand the ohysical principies involved in bringing thesevariables Under control.These problems and their practica solutions have of coursebeen discussed by flute players(and \Vriters) for hundreds ofyears. Hotteterre4 in ca. 1700.Quantz5 in 1752 and Boehm 6in 1871 all wrote on the maer, andtheir books are well worth the attention of a modern flute player.Quantz's book is an indispensable4. Jacques-Martin Hotteterre, Rudiments othe Flute, Recorder and Obo (1707), trans.by P.M. D o u ~ l a s (New York: Dover Publ.,1968).5. Johann Joachim Quantz, On l ~ y i n g theFlute (1752), trans. by E.R. Reilly (London:Faber and Faber, 1966).6. Theobald Boehm, The Flute and FlutePlaying (1871), trans. by D.e. Miller (NewYork: Dover Pub ., 1964).

FEBRUARY 1974/THE INSTRUMENTALIST 7


2/5

tiite etsWslfriii


3/5

Fig. 1. Lip positions for various notes PP mfover the compass of the flute from low C to high G.In the first column the fine wire shows the edge of the embouchure hole.

The reason for the variation inlip pening s largely so that theupper notes, with high blowingpressure, will not be too loud inrelation to the lower notes. In allc:ases the lip opening is approxi-mately an ellipse with width about15 times i ts height. The dimen-sions vary typically as shown inFig. 2. Range of blowing pressures usedfor differenf notes and dynamic l e ~ e l sbya selection offlute players.Measuredcurves are always of nearly the sameshape and lie within the shaded regionshown, most players within the darkregion.

20.- --------------:-----:

'~15

.:

c

Figure 4. It is likely that sorne ofthe indiv iduality of flute tone isdetermined by the exact shape ofthis lip opening and this in turndepends on lip shape and theposition of the teeth. We shouidtherefore treat these curves onlyas a guide and not try to reproducethem blindly.Control of Pitch

The flute is nominally built toplay a perfect. equally temperedscale and we might say th { it isFig. 3. Lip-to-edge jet lengths used bya selection of flute players.

l

being played 11in tune when thishas been achieved. In fact, however, string players tend to playin Pythagorean rather than eitherjust or equally tempered intona-tion, pianos have stretchedtaves, and any sort of ensembleplaying always involves contin-uous adiustment of intonation toFig. 4. Dimensions of the lip opening asa function of pitch and dynamic level.The opening is in all cases nearly anellipse with axial ratio between 10:1and 20:1.

EE

J.Ql 054:e

EE10

J S.;oc c5 c6 l

FEBRUARY 1974 THE INSTRUMENTALIST 59


4/5

FlutsFactsWslfrld Kujli/1 Contribu tng i;d tor

produce a musically satisfactoryresult.The theory and experiment outlined by John Coltman shows thati a flute player simply blowsharder, without making any otheradjustment, the pitch of the notesounded will rise. Essentially, thisis because the airstream vibrations always reach the cuttingedge a little early and effectivelyhurry the pipe oscillation. We

might expect the same result forthe same reason if the lips arepushed forward to shorten thejet length, but often just the opposite occurs and the note getsflatter. Depending on exactly howthe push forward is achieved, theprotruded lips can cover up moreof the flute embouchure hole, resulting in a flattening effect whichmore than compensates for thesharpening produced by the shorter jet.The usual methods taught forflattening a note (turn the fluteinwards, blow clown into it, orlower the chin) or sharpening one(turn the flute out, blow across it,or raise the chin) probably produce the desired effect on thepitch by l t e ~ i n g the amount ofembouchure hole uncovered. f theplacing of the flute against the lipis also changed, or vares from oneplayer to another, then this affects not only the general pitchlevel but also the timbre of thesound.Control of Volume

It is not possible to change theloudness of playing a flute withoutaffecting the timbre of the sound.This may not be a disadvantagebut it should be, at least to sorne

60 TH INSTRUMENTALIST FEBRUARY 974

extent, under the co.11trol of theplayer. For a simple change involume, the goal should be to keepthe timbre as nearly as possiblethe same.To change the loudness of thesound produced by the flute wemust change the amount of energycommunicated from the jet, andthis means a change in the totala ir flow. This could be producedeither by changing the blowingpressure, while making appropriate lip adjustments to keep thenote in tune, or by altering thesize of the lip opening to changethe cross sectional area of thejet.

Measurements show that whenthe player attempts to keep a similar tone quality for loud and softnotes, he uses a constant blowingpressure and vares the size of hislip opening. A clear set of photographs to illustrate this is shownin Figure 1 and the general rangeof variation is depicted graphically in Figure 4.

It is interesting to note that in nocase for a good flute player with aclear tone) does the width of thelip opening greatly exceed that ofthe embouchure hole (about 14mm), nor does the height of thelip opening exceed about 1.3 mm.Obviously an opening wider thanthe embouchure wastes air andprobably produces unwantedbreath noise. We might have expected, however, that a jet morethan 1.3 mm in thickness wouldhave had more power and produced a louder sound than a 1.3mm jet. However, the larger lipopening and thicker jet simplyproduces a dull mf sound withlittle intensity. The reason forthis is that the thicker jet is lessresponsive to the tube vibrationsand so produces a tone with lessharmonic development.

It is. of course, possible to reduce loudness still further bvreducing blowing pressure. butthis is more likely to affect theupper harmonics than the fundamental. It is more useful. therefore. to consider this as a controlled change in timbre.Control of Tone

The last aspect of flute technique that I wish to discuss isthat of tone or timbre. The general sound of the flute is deter-

mined by its design, particularlyby the diameter of the tubing used.The note D5, for example, soundsvastly different when played on astandard flute, an alto flute or ona piccolo. Even for a given noteon the flute at a set dynamic level,there is still a good deal of subtlevariation possible. Let us leave adiscussion of vibrato until laterand consider just the character ofthe steady tone.

The main factor determiningtone quality is the relative strengthof the harmonics present in thesound. The flute is generallythought of as producing a ratherpure sound with few harmonics,

and many of the standard textsshow oscillograms or sound spectra to indicate this. It may be thatstyles of flute playing havechanged considerably in the lastfifty years or that analysis techniques have improved. but ourideas on this point require revision. Analysis of a standard richflute tone on low notes shows arelatively weak fundamental toneand considerable strength in harmonics up to at least the tenth.On low C4 played strongly, infact, the second, third. furth andfifth harmonics are all strongerthan the fundamental. The situation changes as we go to highernotes. In the middle octave the fundamental is strongest with harmonics up to about the sixth beingimportant; while in the top octavethe fundamental is clearly dominant and onlv harmonics up to thethird occur in appreciable strength.

Theoretical analysis and experimentation with adjustable organpipes shows that, for a pipe ofgiven diameter, harmonic development is favored if the air jet isthin and has sharp boundaries andif the jet travel time is less thanthe half-period value we discussedbefare Harmonic development isalso favored if the pipe is sounding strongly rather than weakly.When applied to the flute thismeans that to produce a tonerich in harmonics we should usea lip opening which is a long thinellipse, rather than a wide ellipseof equal area. We should tend touse a slightly lip-forward positionto reduce the jet length, and weshould use high air pressure,though stopping short of overblowing in both these latter ad-


5/5

justments. Conversely, a longerthan normal jet length, a rounderlip opening and a lower blowingpressure will all reduce the har-monic development of the sound.

Sorne of these parameters varyconsiderably from one flute playerto another, and this contributesto the individuality of flute tone.For a given performer, blowingpressure, lip opening shape and,to a lesser extent, jet length pro-vide the major possibilities foradjustment. Stretching the lips tothin clown the elliptic opening andincreasing slightly the blowingpressure yields a more intensesound, while a relaxed embouchureand a lighter blowing pressuremakes for a lighter, quieter tone.All these adjustments can be overdone, of course, and too round alip opening or too low a windpressure leads to dullness, whiletoo tight C:lll embouchure and toohigh a blowing pressure gives astrident sound. Within these limits,however. there is room for quite adegree of suhtle tone variation.

Another component of the tonewhich needs to be under controlis the amount of breathiness orair noise oroduced. This is mostpronounced. of course, during thearticulation at the beginning of anote and here it is a necessary com-ponent of flute sound. During thesteady part of the note, however,we want to reduce it to the lowestlevel possible. Too thin a jet encourages wind noise, as does strayair not reqching the embouchureedge. In general. therefore, weshou1d strive for a regularly shapedlip opening which is less in widththan the flute embouchure hole.Again, there is a good deal ofindividualitv in this and we shouldnot seek undue uniformity.Vibrato

Vibrato is not strictly part ofthe steady sound, but it is usefulto discuss it briefly here. Thereare actually three variations ofsound involved in a normal vibrato - a slight variation in pitch,a variation in loudness, and a var-iation in harmonic content. For astring instrument, the pitch varia-tion is the main feature of the vibrato; but for wind instrumentsthis is of less importance andvariations in loudness and har-monic content predominate.

Measurements of blowing pres-sure in the mouth suggest thatthis fluctuates_ by about 20 percent above or below its mean valuein the course of a tvpical notewith vibrato. The resulting fluctuation in sound level is typicallysomething like a factor of two, butmuch more important is the fluctuation in tone quality. Thus, whilethe amplitude of the fundamentalvares by onlv about 20 percent,the amplitudes of the third. fourthand fifth harmonics. which are asstrong as or even stronger thanthe fundamental. may vary bymore than a factor of six, following the pulse of the vibrato. Theassociated pitch variations are, incontrast, very small. We shouldtherefore characterize a flute vibrato as predominantlv a timbrevibrato. rather than either a pitchora loudness vibrato see Figure S .

ConclusionIn this article I have discussedsorne of the basic factors influ

encing the production of soundfrom a flute. What I have notdone, however, is to discuss themuscular adjustments which aremost appropriate to control thesefactors, and this is, of course, ofeven greater importance to theflute player.

It seems that there is sometimesquite a difference between whatwe think we are doing in the mat-ter of embouchure control andwhat we are actually accomplish-ing. In itself this is of no impor-tance at all unless we try to enforcea uniform approach on all ourpupils. I hope, however, that arather deeper understanding ofwhat we are trying to do may helpus all.

C 6 Hz C 5 4HzFig. 5. Harmonc structure and vibrato on notes C4 ond C5 The relotiveamplitude and time variation of each of the first six hormonics is shown, thepulse of the vibrato being 5 per second in each case.

FEBRUARY 197 4/THE 1NSTRUMENTAllST 6

acoustical principals of flute technique

Documents