The theory of sound / by John William Strutt, baron Rayleigh,...

Source gallica.bnf.fr / Ecole Polytechnique

Rayleigh, John William Strutt (1842-1919 ; 3rd baron). The theory of sound / by John William Strutt, baron Rayleigh,.... 1877-1878.

1/ Les contenus accessibles sur le site Gallica sont pour la plupart des reproductions numriques d'oeuvres tombes dans le domaine public provenant des collections de la BnF.Leur rutilisation s'inscrit dans le cadre de la loi n78-753 du 17 juillet 1978 : *La rutilisation non commerciale de ces contenus est libre et gratuite dans le respect de la lgislation en vigueur et notamment du maintien de la mention de source. *La rutilisation commerciale de ces contenus est payante et fait l'objet d'une licence. Est entendue par rutilisation commerciale la revente de contenus sous forme de produits labors ou de fourniture de service. Cliquer ici pour accder aux tarifs et la licence

2/ Les contenus de Gallica sont la proprit de la BnF au sens de l'article L.2112-1 du code gnral de la proprit des personnes publiques. 3/ Quelques contenus sont soumis un rgime de rutilisation particulier. Il s'agit : *des reproductions de documents protgs par un droit d'auteur appartenant un tiers. Ces documents ne peuvent tre rutiliss, sauf dans le cadre de la copie prive, sans l'autorisation pralable du titulaire des droits. *des reproductions de documents conservs dans les bibliothques ou autres institutions partenaires. Ceux-ci sont signals par la mention Source gallica.BnF.fr / Bibliothque municipale de ... (ou autre partenaire). L'utilisateur est invit s'informer auprs de ces bibliothques de leurs conditions de rutilisation.

4/ Gallica constitue une base de donnes, dont la BnF est le producteur, protge au sens des articles L341-1 et suivants du code de la proprit intellectuelle. 5/ Les prsentes conditions d'utilisation des contenus de Gallica sont rgies par la loi franaise. En cas de rutilisation prvue dans un autre pays, il appartient chaque utilisateur de vrifier la conformit de son projet avec le droit de ce pays. 6/ L'utilisateur s'engage respecter les prsentes conditions d'utilisation ainsi que la lgislation en vigueur, notamment en matire de proprit intellectuelle. En cas de non respect de ces dispositions, il est notamment passible d'une amende prvue par la loi du 17 juillet 1978. 7/ Pour obtenir un document de Gallica en haute dfinition, contacter [email protected].

THE THEORY OF SOUND.

MARS ~a~

TH F

THEORY

OF

SOUND.

HY

JOHN

WILLIAM
FORMEK.Y

STRUTT,
FELMW OF

BARON
TRINITV

RAYLEIOH,
COLLEGE, CAMH)UnOE.

M.A..F.R.8.

VOLUME

1.

&onlton: MACMILLAN 1877

[~~ ~t~~

+ AND 00.

r~trt'ed.]

OEambtitgc:
r)t)M):Y'J' ATT)tKUS)VRM)TV'')t'<H~.

PREFACE.

IN

th

work,

of
endeavour

winch

lent, my a connected should in The

exposition include ~he more times y of the

volume is an instalprsent bas been to lay before the reader of the of sound, which theory of th adv~nces made important Matliematicians and Physicists. object winch 1 have had in view disputed Ly those time of th many are to be found to competent most valuable in scattered

th

modem

importance will I tinnk, not, be At th prsent judge. contributions to science

onjy

and penodicala traiMactions of socletles, in pubMied vanous of th worid and parts in several and languies, arc often inaccessible to those practically vvho do not to hvc in th happen of large neighbourhood public hbraues. In sucli a state of things the mechanical to impedimonts entail an amount of unremunerastudy tive labour and hindrance to th consquent advancoment of science which it would be dimcult to overestimate. Since c'~)~
no complte

the

wcH-known

Article

on

Sound

in

the

.E~-

~)~M~(~ work is trcatcd Prof.

lias

John by Sir been publislied

Herschel in

(1845), wilich tlio

suhject death
of

of

jnatl-cmatically. Donkin tlie scientific

attainments of mua:c

onc a

w!)oso
{n-ctic.-d

Ly th promature worid was deprived

in quahned combmation hifn in

mathcmatical kr~Judge

with

VI

PREFACE.

manner to write on Sound. special The first part of his Acoustics little jnm~ tbsm (1870), a fragment. though is sumcient to shew that would l~ave been unmy labours had Prof Donkin lived to complte necessary his work. In tlie choice of topics to be dealt with in a. work on Sound, 1 have for th most part foUoAved th exemple of my predecessors. To a great extent th of theory as commonly Sound, covers th same ground understood, as th of Vibrations in gnral theory some but, unless limitation were th consideration admitted, of such subas the Tides, not to speak jects of Optics, would have to be included. As a gnral ruie we shall confine ourselves to those classes of vibrations for which our ears afford a made and ready sensitive instrument of inwonderfully Without ears vestigation. we should care much hardiy more about vibrations than without care eyes we should about light. The volume present includes on th vibrachapters tions of systems in gnerai, in which, 1 hope, will be some recognised of treatment and novelty fol. results, lowed by a more detailed consideration of special systems, such as stretclied bars, and strings, membranes, plates. The second of which volume, a considrable is portion will commence already written, with acrial vibrations. best thanks are My due to Mr H. M. Taylor of Trinity Collge, who bas been Cambridge, good enough to read th proofs. assistance By his kind several errors and obscurities have been and th eliminated, volume has been rendered generally less imperfect than it would otherwise have been. Any corrections, which my rcaders ciated.
TEttUNO .~n7, PLACE, 1877. WmiAi.,

or suggestions for improvements, me will be may faveur highly

witli apprc-

CONTENTS.

C'HAPTER
1-27 Sound duo to Vibrations. Finite

I.
MOE i

of Pitcb. pondent Depmult'a 'Witoatatono'a experiment. and Noisos. Musical motea duo to poriodio vibrations. do la Tour. Pitch dopendont Poriod. upon musical intorval Absolute Table notes. in all Pitoh. Tho parts samo of tho ratio scale. of of perioda Harmonie

of Propagatiou. indoYelooity Sound propagated oxporimonts. in water. Enfoeblomont of Sound Notes by distance. volocity Siren Eelationahip of Cagniard between

to the samo corresponds sca-tes. Diatonio soaloa.

Temperament. Equal Tomperament. of FroqnonoioB. of Notes. Notes Analyais and Tones. Quality overtonos. depandont Resolution of Notes upon harmonie by efu; un. certain. tcnoa to Simple correspond vibra.tiona. aimpla pondidona

Neoossity

CHAPTER
28-4.2 Composition position Theorem. Lissajous' methods termittont of harmonio of two Vibrations Figures, of composition Illuminntion. motions of of like nearly period. equal

II.
18 Harmonie Curvo. Bats. Lissajous' Com. Fourier'a Cylindor.

vibrations

in porpendionlar Bin.ckburu's poudulum. and analysis.

period. directions. Th

Kaieidophone. Optical vibration In. microscope.

viil

CONTENTS.

<HAPTER ~-c.s
Systumswithon~d~rouoffrt~dom. pfriod.

IH.
TA<!t: t.

Indopcndottcu

of

ampiitudu~nd

Frictiomdfurccpruportimudtovotouit.y. J;L'at.S(]uoto )'riuci)do~fSnpurpositiun.

l''))t'cnd\i)))'ati<n)s. H)tpLTp()sitionoff.,rcud~nd M~

fn~vibrutions. thodofDimcnsionH. .f)!-kn<~ standards

Yan<'UHd(~n.c.sofdmnpin(;Stri))Kwit)tL<Hn). Id<~dTtminn-furk. I''('r)iSKivon(.-ar)yp)tr(iio])()M. ofpiich. Scbcibicr'snjutb~dsoftuni; Cofnpnnndi'cndtdmn. l''()rhsdnvuubycluctru. Itcsonfmuo. Guuurftimtlutionforohc

Suh~ib-

Itir'sTonomutM's. t)mK"ft"u. deHt~ooffrueJott). toucu.

!or]tIntt!rn)ptM-.

TM-mHofthotiMondur~orj.;ivcrisetctterivod

C'frAPTER ~
(icnorniixedco-ortiitmtcs. rMua. thoornm. funetion. tion. tion. Itiitml motions. Expression for

IV.
(,7 jMjLontudcncrKy. StftticaithcoHcciprocat ThodiH.si~ftti~n condiArctttKfttif'~

Hxprcsnionforkinet!<:(!)~ur~y. L~'n))HO't!('qutttionn.
l''t'cc\-i).mtion.swt.])out.fri<

ThM))Ho))'sthcorc:r)i. Couxi.stt'ncoofHnMtUinotioxH. Nnrmnico-nrdinntos.

Thch'eoponodHfuXtt~Htttt.ioj~u'y

AnncccHsiottofinertijtmcrottsc.sthefrucpoiods. Tim(;<tt''stfroopcrindi.s nfvntmtinn. types Hy~touM. propnrty. Stotios' StrioK ]')ctcrnn)Mtion thcurom. of

('fHprmf;i))()rL'!Lsc~ti~fM)"i[)dH. ttbtiotntu maximum. Ifypcthut.n'fd ntrinn. Normal nuit Approximntoty fnnctions. initit simpto Conju~to conditions.

n)t Hx)unj))ofr<))u vn!-itth)o dcnsity. of cf.nstnotM tu

M'hitrary

CHAPTER ~17
CascH in nmus pointu denectcd equinbrium wluch of of tho thrce funetiotM Ut;noM)i.tion H(iui)ibriu)u appiied at Y', r of fiquarLi.s. Htrin~H. by n fora

V. <~
Hunu[t!U)oous)y thcorcut Y'uun~'H aro rcjucibio ou tho undft) tu

titeoi-y. tit.n'tcd fruin SyatomH rest as ono point. Ht~rtud 8yntbtUH froui tho

confjgumtiun by an impniHu nt ono point. applicd Syatcma startod from rcst na dcftocted by n. force distributed. JuUuuuifonuiy oico of sw< frictiona) forcoN on tit vibr~tious of ft yntom. Solution of t)m ancrai for freo vibMtiotM. oqn~tioDH Forces, Impres.~d rrinciptc "f tho porsititonco of poriods. Inoxontbto motions. TJ)uo. MceiprocttI ru)!]. to freo vibmtiouB. Applicution Stutemcnt of Mciprooa) t)tcnrcm for itttnnonic forcos. Extension AppHcationH. to cases in wbich thu constitution ofthu systcm for isn.functionuftLpuriud. Equations two dcgrcus of fruedom. HoutH of dctornnmmttti cquittiun. lutct-niittcntvibmtiomj. Marchofporiods incro~od. ttHim.rtin.is~mdu~Hy Heaction of n dcpunduut f.vHtun).

CONTENTS.

Ix

CHAPTER 11814S.
Lnw of extension bJcm rivation Hxpt't.'snious tion. odio. StritiRS for of n string of no)ution for with n, titritig. whoso for nnd Transvurso masH

VI.
pAnx

127
vihrntiona. Solution in cquidistant Pfu'tittI form GcnoDt) Kurtnn.1 initit y in of diffcrentitti simpio of ntotlua of tho pro. D-

is concontrutcd string. gonorat

points. harmonie n stnng of

continuouo y. Hoat extronitios.

quation. mo-

xoJ L)tws.

motion

Mcrsonuc'H

Sonomotur. to suit nrbitt'M'y for ' and motion. l)y tu tm

pcrivibration. Ca.Mo co-ordiplucMn~. of pifuio-

Dctunninat.ion

of eonattuits

ciroumfitttncus. of noriun] by

of pluoknd HtrinR. n<tte!i. Nonntd

Expression.-) cqutttionH of

torms

Strin(; impulso.

oxcitod rroblem CoBtptn'ison

cxcitud thoorou. Htiing YoHug'a forto Friction j~'oport.ionni strinH. Iibritttn to of l'oriodic force tLeory. of tho cxtremit.iea. yiuldiu~ a nnitu loud. on Correction curved fur stretchod

uppliud Proof for

vulooity. nt onu uf

with

oquiduu EGuets strin~. of tho

point. Fouriur't)

Modificationa thcorcm. of tho violin caso of of

Striuss

ProUm rigidity. (iurfaeca. Solution for of donsity.

Correction iiphoro. for and LiouYiDo tdonj~ tinito an uniimitod

irrcguJaritioa of vtu'inbio Positivo )L uxod

TheoMinH

Sturm Wt).vcs

n strin(~ Htring. at

donsity. rropng~ioa fmd wavos. nc~~tivo Dduction of

Stn.tiona.ry solution friction. for

Vibrations. strit)R.

Hcnootion Grn.phiod

point.

mothod.

Progressivowawwitli

CHAPTEK

Vif.

1491~
Ciftsxincttt.ion dimd tion ona of tho vibrations Numorieal frco nnd at both onc lixod. Ltuof Bars. valuos uodH. l!oth with DiuM'cntin! of tho vibt'jttiottH. for end a Lar fret), of quation for constunta of tilution InUuoncu n.ttMued. longitu. Solustel. fur a Lar with Inad. for fortora.nd torfor

188

Dduction ends ItH'gu iixcd. Joad

of Mniull Corrcctiou

Solution In.tcft.I Hioniti tiiomd

for problom SavM't'H motion. vibrations. wa.'vos.

"Hou

Comp~risou

Differentini rauqno." quation uf vclocities of longitudinal

CHAPTER
IGO1U2 energy of bcndhtg. Expression Termimtl Conjuguto
sq~n'CH.

Vni.
201 for conditions. property of of r tho in kinctio energy. G encrt normal tcrms Drivation solution functiona. of not'nml cufor

PotcutiftI

of diSercutitd tt hnrmouic YatucB

equ)ttion. vibrittiou.

ufintrKratcd

ExprcsHi).])

CONTENTS.
l'AUI~

ordinatos.
to suit

Normal initial

equatious

of motion.

Dtermination

of constant))

conditions. Caso of rod etartod I~od start~ bv a b)ow. as dofloctod ~t~I ~~c. by lu CL.~m ~os tho BorioH of normal funoticus coanos tu coj.vcro. Form of H~ norma! hmettons li(:u-fiue bur. Lfnv9 of Jopotj.Icuoo of frofjuoucy und tLiokou )ongt)i noM. Caso whou both oudH ~rc clampod. Normal fuuctions for ,) bar. clampod.Ireo Caleulaticu o puriud.s. of pitch. CompnrinonH Disoussiou of tho gravost modo of vibration of a freo-freo bar. Threu nodos. Four ~oJoa. Gravost mojo for clampod-froo bar. Position ut nodos. bar. Supportcd Calculation of poriod for clamped-froo bar from Lypothetleal Solution of typo. for n bar with a loaded problom ond. Euuct. uf adtUtious to a bar. lufluonco of irrogularitios of donsity. CorreottOM for rotatory iuortia. L:oots of functioua dorived iinoarly from normaJ fuuetioM. Formation of ~uatiou of motiou ~heu thoro is por. Mauout tousiou. trmiual Spoeial couditioua. Itosultaut of two trains of wavcs of iic-iirly cqua) poriod. Fourior'H solution of problom for ini. nito har. from rest

CHAPTER

IX.

1~213. Tension of motion. Equation Fixod reotangular lonudary. for Expression ~aud iu tenus of normal co-ordiuatos. Normal of ouations vibratiou. of improssed forces. Examplos for Frequoncy an olongatod rectangle ou tho shortor dpends sido. Casoo iu maitiJy whioh difTerout modes of vibration havo tho samo Dorivod poriod. modes thence Effeet of 6li(;ht irrcgulanties. arising. An irregu!arity may rontovo nidoturmiuatonosa of normal modoa. Solutions applicable to a triaugle. of tho Honorai Espro~ion diilorontml eqnation by polar co-ordm~tes. Of tho two functions, w),idt oceur in tho solution, ono ia cxcluded by tho condition at tlio polo. for Bossel'a Expressions functions. Formutm theroto. Tublo of tho rdating first two functions Fixod eiroilar boundary. of tho normal Conjugato proporty functions without restriction of boandary. Values of integrntod Exsquares. for T nnd F in tcnua proHMoun of normal functions. Normal oquations of vibration for ciroular mombraue. easo of froo vibraSpoci.d tions. Yibratioua duo to a harmonie force distributed uniformty Utohos of tho varions tonoH. Tabio of tlio roots of Bosscl'o shnpto functions. Nodal Fiur~. Circular mombrano with ono radius fixed. Bessel's B onctions of frnctional ordcr. Ejloct of sma'I lond. Vibrations of a mombrano whoso boundary is approximatoiy ciroutar. In many casos th pitch of a mombrano mny bu calculated from tlio aroa alono Of aU .nombraues of equal aroa t)Mt of oireular form l.M tlio gravost l'itch of a mcmbrano pitch. whoso boundary ia au eDipso of smalt ceeHntricity. Motliod of obtai)iii)g limits in casos that oumot bo dealt witli rigorouf3ly. of fruqueucioa Comp~rison iu varions ca.sc.s of mcmbraues of eqna) arc.a. of tho probion.. Histury Bourh'ot'8 oxperi. aonta) invostigfttiouB. of a motubrano.

CONTEN-TS.

XI

CHAPTER X.
RAOP 214-235 Vibrations of PIatos. of normal Potontial Enorgy of Bending. Transformation conditions. to polo, polar of 5~.

Superuoial

diiorontial

proporty Form of gonorni solution continuons tho for a froo ciroular mining poriods with observation, Comparison mdii Kivo riso to boats. cdgo. critieiamo. Gonoralizution Disturbn.uce Hupportod, Mn.tl.iou's nnguhn-

equation. functions.

Dou.ndary Transformation through p!nto. of nodal

Conjugato co-ordinates.

doterEqnations EirohhoC'a catouhtions. cirolos. Cnso of Irreguln.nties or cJampod, of proUom.

of solution. figures.

of Chiadni's

Hifitory

DoetfmguiM with aupportoJ phtto Itoctedgo. plato with freo edgo. couditionH. Ono Hpocial cnso Boundary iH funonablu to mfttttomaticfd (~ = 0) tro~tmont. of codai Investigation figures. WItcntatoue'H application of tho mothod of Hnporposition of CompariMU Whoat~tono'f. wit]. thoso liguros to n reaUy n.pp)io~!o in tho cnso = 0. pMo Gravost modo of a squnro plate. Caiouhttion of poriod on hypothotica! Nodal type. from ~igurcH inferrod considor. atlona of symmetry. Hoxngon. hotweon circle nnd squnre. Comparison Lnw and thicknoas. connooting pitch In tho cnso of a elfunpod odgo of tho boundary nny contraction raisos tLo pitch. No gravest form for a free plato of givon aron. In similar tho poriod is as tho linoar plates dimension. Whoat.stono'a on wooden exprimenta plates. Knig'a Vibrations oxperimontN. of cylindor, or ring. Motion as tangentinl woll as normal. Bolation betwoon and normal tangoitial motiona. Exfor Mnetio prossinna and potontial of vibration. nergies. Estions of tonos. L'requoncios with Comparison Chiadni. friction Tangential excites motion. tanguntiat Exprimental vrification. Bats duo to irregularities.

CIIAPTER

I.

INTRODUCTION.

1.

Tim

sensation

of

sound

with parab]e th relation or

any of our other between a sound questions to thc be Whcn no

is a thing sensations. and as a colour connected

s:M ~e~eW~, not comNo one can express or a smcil. this Directly must subject and hcaring; thcrefore with thc to thc

ail indirectiy, comc for decision from infci' it that thct'c ail car. which can

with

car, appea!.

t!tG But

of organ wc are not arc discovercd of

acoustical

unassistcd phenomena

investigations once wc have th

conducted

constitute

foundation

arc in great mcasurc transferred plorations within thc dominion of th pi-mciples of Mcchanics. laws arc in this way ai'rivcd at, to which the sensations canuot but conform. obscrvatioo. Very cursory bodics arc in a statc sounding nomena of sound and vibration bell or string is touched vibrating 2. often succs and

sound, to another nc!d

physical our exlying

ImportMit of th car

to tha.t

shew thc

that

of vibration, are closcly

connected.

p)ica, WIicn

th sound cea~cs by the finger, at th same moment is damped. tha.t thc vibration But, in order to affect th sens of hearing, it is not enough to have a vibrating t!icre must also be an uninterrupted communication instrument between thc instrument and thc car. A bcll rung in ~ac!<o, with proper rcmains sounds without th prcautions inaudible. have break a from of the to th prevent In th air of communication th of motion, howevcr; atmosphre, of conveying thcin sources to

univcrsal th ear.

vehicic, capable most constituted var)ous~y

rccesses 3. Thc at

is y

jn'cd

is net instantancous. passage of sound a distance, a very perceptible interval

a g)in th spartes 1

Whcn

2
report souud from the flash.

INTRODUCTION.

[3.

This

experiments French in 1738. Academy, and th rctardationof th reports at different distances ohscrvcd. Th principal is to revo-se prcaution tlie -necessary alternatcdy direction along which the sound travels, in order to cllminatc tho influence of tlie motion of th air in mass. Down t!ic wind, for sound travels instance, to th carth faster than its reJativeJy proper rate, for the velocity of thc wind is added to that proper to the propagation of sound in still air. For still dry air n.t a of 0"0., thc French temprature observera found a velocity of 337 metres Observations of tho samo per second. character werc made by Arago and others in 1822 by th Dutch Moll, physicists van Beek and Kuytcnbrouwer at Amsterdam and by Bravais Martins between thc top of the Faulhorn and a station bclow and by others. Th gnerai result bas been to give a somcwhat lower value for tbc of sound-about 332 mtres velocity per second. Thc effect of altration of temprature and pressure on the of sound will be best considered propagation in connectiou with th mechanical theory. members of the of observation, that consequence within wide th velocity of sound is independent, limits, or at least very ncarly of its and independent, also of its pi tel). intensity, Wcre this a of music otherwiso, quick would piece be hcard at a little distance confused and discordant. hopelessly But when the disturbances are vcry violent and abrupt, so that th altrations of density air, the 5. concerned simplicity An claborate arc comparable of this law may sries with th whole from. density of the be departed 4. It is a direct

in traveUIng from th of the nash duo to th finite Th first accuratc negligible.

rcpresents gun to th velocity

the observer, of light

time

occupied by the rotardatinn

bcing altogether wero mado by somo Cannons were nrc-d,

of experiments on tlic propagation of sound in long tubes has been madc (watcr-pipes) by Rcgnault\ He adopted an automatic similar in principle to that arrangement used for me~suring th speed of projectiles. At thc moment when a ptstol is fired at one end of tlie tube a wire an electric conveying current is ruptnrcd Tins causes th withdrawai by thc sliock. of a tracing point which was previonsly a line on a marking revolving drum. At tho furthcr end of thc pipe is a stretched membrano so that whcn on th arrivai arranged of the sound it yields to th ~MofrM <?<: rjca(~;);. ~e ~-/tc< t. xxxvn.

5.]

VELOCITY

OF

SOUND.

which was ruptured of th the circuit, tho passage during impulse, At thc sa.mc moment soun< i3 rccumpietfd. tho tracing point Tho blank on tlic drum. correfaits back space loft uumarked the joumcy, to thc thuc occupied by th Sound in t~aking sponds and, wltcn th it.
membrane

motion Tho
is

of th length fouud

drum ofthe

dctcrmining and the cxperimcnts pendent to 1'100. j innucuco G.

is known, givcs the hctwccn th journoy

mcasurcmcnt.

means first In

of wiro

of

the velocity th dl~meter Tho hc greater

by of sound of in the is

direct

thcsa

to hc not quitc indcappcarcd vn.)'Icd from 0'108 pipe, whieh to friction, whose

diso'cpancy

would AIthough,

duo perhftps smaller pipes.

in practice, air is usually the vehicio of sound, otiicr and solids are equally of conveying capable gases, liquids a direct mcasureit. In most cases, I)owever, th means of making are wanting, and wo M'e not yet in ment of the velocity of sound tlie a position to consider water tho same diniculty S Colladon Lake second arriva! of simultaneous t, station of th ami Sturm with Geneva. indirect does not methods. occur. th propagation of a bell at But In in thc th one year of sound station caso of 182G, in thc was

investigated Tlie striking a nash the of

mcasured

The observer at a. gunpowder. interval between tho flash and the beneath sound In

Itis car to a tube carried sound, applying of 8C., th velocity At a temprature of th surface. water was thus found to bo 14-35 metres per second. 7. Thc

pretty is connectecl taken

of sound conveyancc by solids may bc IHnstrated due to Wheatstone. One end of a metallic experiment with tho sound-board

by a wiro

through where building, nancc-board

of a pianoforte, and th other or floors into anothcr of th th partitions part would be audible. If a resonaturally nothing with th the sound

as a violin) bc now placcd in contact on th is easily heard, and wire, a tune piano p]ayed from th resonance-board. seems to cmanatc (such 8. rapidity amuunt In an

of sound falls off with grcat open space thc intensity from th source increases. saine as tho distance Th of motion bas to do duty over surfaces ever Increa~ing as the

that confines the sound will of the distance. Anything squares ttte falling off of intensity. to dimini.sh Thus over th flat tend can'Ies furthcr broken of still than over surface watcr, a sound ground is still thc botter; corner but between the most a smooth crtcctive and a vertical wall pavement ofaU is a tubc-likc enclosure, 1S

INTRODUCTION.

[8.

prevents Tlie use of speaking tubes spreading altogether, to faciMtate communication between thc dirent parts of abuHdir)<r is wcll known. If It were not for certain crfects (fnctionat and .other) due to th sides of th tube, sound might CD be thus conveycd with little loss to vcry great distances. 9. w!uc!t
Sounds

which

Bcfore is of

furUicr wc must consider a distinction, procecding grcat not frce from unportance, though dimculty.
bc ciassed as musicn.! a)jd unmusica] thc former for

may

bc caHed notes and titc may lattur noises. Tho(,, extreme cases will raiso uo dispute; one rccngniscs thc every diffrence betwecn th note of a pianoforte and t)ic ereaidng of n shoo. But It is not so casy to draw t]ic line of Li the sparation. first place few notes arc frcc from a!i unmus:c:d accompanimcnt. Wit)i thc hissing organ of th wind as it pipes especially, escapes at thc mouth bc Iteard beside the proper may note of tlie pipe. And, second]y, of a musical many noises so far partage character as to hve a definite T!tls is more pitcb. in a easily recognised tite common sequence, giving, forexampJe, chord, than by continued attention to an individual instance. Th experiment may Le made corks from bottles, by drawing tuned water previously by pouring into them, or by do\vn on a table sticks of wood of suitable throwing dimensions. noises are But, somctimes not entirely although and notes arc usually not unmusical, Is quite free from noise, thcre no diniculty in recognising which of th two is th simpler phenomcnon. Titerc is a certain smoothness and continuity about tho musical note. Moreover a variety of notestogether bysounding for example, a number by striking of conscutive simultaneousiy on a pianoforte-we keys obtain an approximation to a noise; while no combination of noises could evcr bicnd into a musical note. 10. We arc thus led to give our sounds. Thse attention, in ttic first instance

convcaicnco

to musical mainly themselves an'angc naturall in a certain order to pitch-a which according ail can quality to some extent. appreciate Tralned ears can recognise an enormons of gradationsmore .numher than a thousand, within probably, the compass of the human voice. Thse gradations are of pitch like the degrees not, of a thermometric without scale, special mutual relations. note as a starting Taking any given point, musicians can single out certain which others, bear a definite relation to thc first, and are known as its octave, fifth, &c. The di~i-ences of pitch corresponding arc cal!ed and arc intervals,

10.] spokcn
"horov.'r

piTcn. of as always th samc

lli th

5 same
a. note

for th
Hcale,

relationship.
'~d ita. octave

Thus,
arc

th~yMn.ynccm'

It will be our object later sep~u'ated by </tc ~~o'uf~ of ~te oc~~e. so far as it can be donc, tho origin and nature to cxplain, of the consonant but we must now turn to consider intervals, th physical aspect of tlie question. are produced it is naturel to suppose by vibrations, that tho simpler sounds, viz. musical to ~e/~o~'c notes, correspond that is to sa.y, vibrations which after vibrations, a certain interval themselves of timc, called th per~~ with perfect regularity. repcat And this, with a limita-tioM prcseutly to bo notioed, is true. contrivances bo proposed to illustrate Many tlic may of a musical note. One of th simplest is a revolving gencratln w)icol whoso milled is presscd a card. cdge Each against as it strikes the card gives a slight projection tap, whose regniar as the whee! turns, produces rcurrence, a note of definite pitch, the fM scale, 7't'A-t'yt~ MM?-casea. But th most velocity of p't~b?!. instrument for the fundamcntal uppropriatc on notes experiments is undouhtediy tlie Siren, inventcd de la Tour. It by Cagniard of a stiff dise, capable of' revolving essentially about its and pierced with one or more sots of holes, centre, at arranged round th circumfcrcnce cqual intcrvals of circles conccntric with th dise. A windpipe in conncction with bellows is presented to th dise, its open end perpendicularly to one of bcing opposite th circles, which contains a set of holes. When th bellows are the strcam of air escapes worked, to tlie frcely, if a hole is opposite end of tlie pipe but othenvise it is obstructed. As th dise turns, a. succession of puffs of air escape when it, until, the throngh is sufncicnt, into a note, vclocity whoso they btond rises pitch with the rapidity of th pun's. continually \Vc shall have occasion later to describe more claborate forms of th Siren, but for our immdiate 12. purpose One th prsent simple arrangement facts in th will sunice. whole science is cousists 11. Since sounds

of th most

important

that th pitch of a note dpends cxemplincd by tlie Sirennamciy, Tho size and shape of th holes, upon th pcriod of its vibration. the force of tlie wind, and other. lments of tlie problem may be but if th number of puffs in a given varicd time, such as one romains so also does the pitch. second, We may even unchanged, with wind altogethcr, and produce a note by allowing dispense th corner of a card to t~p against the cdges of the holes, as they

INTRODUCTION,

[12.

will still be th same. Observation of othcr pitch of sound, such as vibrating consolids, leads to the samo th difficulties arc often such as to render clusions, though rather rcnned mothods. necessary exprimental But in saying that there pitch depends upon. period, revolvc sources tho lurks as in it the variable an ambiguity, us will lead which to a dcscrves attentive of grt point is periodic in any time -r, it is also quantity timos a recurrence within 27-, 3, &c. Conversely, not exclude a moro reourrence r, docs rapid which that are of r. aliquot parts a vibration really recurring the tho It in th consideration, If importance. periodic a given within appear (for by by a

period periods

would time

according!y

as having example) may be regarded laid down as produciog tlie lawjust T. Thc force of this consideration defining as tho

-r, and pcriod a note of the pitch cannot be entircty

~r therefore defined evaded

th least time to bring pcriod rcquired In tlie first place, th necessity of such a restriction is rptition. in itsc!f almost sufHcient to shcw that we have not got to th root of the matter fur although a right to th period r may be dcuicd to a vibration itself rigorousiy within a time ~T, yet it rcpeating to a vibration must bc auowcd that may differ indefinitely little thcrcfrom. In thc Siren

by about a

that in one of thc cxperimcnt, suppose ch'cles of holes containing an cvcn number, hole is every alternate th arc of the circle by the same amount. The disp]accd along dplacement may bo made so small that no change can be detected in tlie resulting but the periodic time on whieh th pitch note lias bccn doubled. And secoudly it is vident from th dpends nature of pl'iodicity, tliat th superposition on a vibration of period docs not disturb the period r, T, ofothurs having pcriods ~T, ~T.&c., while be supposed that th addition of th new clyet it caniiot ments bas left thcqualityofthe sound unchangcd. Moreover.sinco thc pitch clcmcnts 13. between natural affectcd hy their of the sliorter periodswere Thse th numbers. considrations
notes whose

is not

how do we kuow that prsence, not tbercfromt)ie beginnin"'? rcJations of th

lead us to expectrcmarkable are as th reciprocals periods

can bc easicr than to invcstigate th Nothing two circles of holes, the <tucstion by meaus of tlie Sirot). Imagine inner and th outer twice as number, containing any convcnicnt TIien at. wfiatcvcr th dise may turn, th period of many. specd the vibration the first set will necessarily engendcred by blowing

13.]

MUSICAL

INTERVALS.

be th double experiment th relation ?M<e method

of that the two

belonging notes are

to

th to

second. stand that

On to

found

the making cach other in

of octaves; and its octave, the ~'c~c~/ of experimenting the interval octave and a

we conclude

corresponds up of an

is of vibration shews, that to th ratio of periods 3 1 known to musicians made asthe<we~ to th ratio of 4 1, th double fifth

in passing from any doubled. A similar

mado up of two octaves and to th ratio 5 octave; 1, th interval of the fifth and a major </Mr~. In order to obtain tho intervals and third the ratios must 3 2 and 5 4 be made thcmselves, respectively. 14. stand From those experiments in a fixed that if two notes appears at what then, no matter relation, their are in a may bo situated, periods of th relation. The same characteristic of 2 to vibrations 1 is thus combine it

to one another

part of th scale they certain constant ratio

or tho number may be said of thcir /?'e~Me?tc~ winch in a given Th ratio time. they excute characteristic of tho octave intcrval. If wo wish two Intcrvals,for from a given instance, starting a step of an octave and then another of a fifth direction, the corrcspondine ratios must be compounded

note, to take in th same

is represented part of an octave by the ratio !V2': 1, for tins is th stcp which twelve times leads to an repeated octave abovo the starting If we wish to have a measure point. of intervals in th proper take not the charactersense, we must istic ratio only, will ucasurcs of th compouonts. 15. sidered From the of tliat ratio. itself, but th logarithm Then, and then the mcMuro of a compound intcrval bc the SM~ of th

Tlie

twelfth

intervals

known above, othcrs difference of an octave and 3 ratio This process 2~=~. th octave is called ~uer~M:~

conoctave, fifth, and third Th to musicians may be derived. and ha~ the a fifth is called a fourth, of th of it. subtracting By inverting an the interval major from third

A single word to donoto tho numbor of vibrations oxccuted in tho unit of timo is indi~ensabio: I know no butter than froquoncy,' which was nsod in this sonso by Young. Tho sMto word is omployod by Prof. Everott in bis excellent odition of Doscbanol'a ~atw<t! P/(t'!osop/t~

8
we obtain th minor

INTRODUCTION.

[15.

of a major Again, by subtraction third from a fifth we obtain aud from this by th minor third; inversion tho major table sixth. The following exhibits side by side th names of the intervals and the corrcsponding ratios of sixth. frcqucncies Octave Fifth. Fourth. M~jorThird. MiuorSixth. Miner Third. M~jorSixth. Thcfjo are ail thc consonant intervals 2 3 4 5 8 G 5

witttin thc comprised limits of th octave. It will be remarked tliat tite corresponding ratios are ail of ~M~t~ whole numhers, and expressed hy means t!tat tliis is more particularly th case for th moro consonant intervals. The notes whosc frequencics
AM~M~M,

arc and

to violinists, they may ail bo obtaiued from the samc it lightiy with the string by touching whilo th bow is drawn. imgcr at certain points, Tlie establishment of th conncction between musical intervals and dfunte ratios of frequcncya fuudamcutal point in Acoustics -is duo to Mersennc It was indeed known to th (J63C). Grceks iu what ratios tlie Iougtlis of strings must bc chaagcd well known in ordcr but Mcrsenne duntourifth; strated tlie Jaw connecting thc length of a string with the ponod of its vibration, and madc thc first dtermination of the actual rate of vibration of a known musical note. octave 16. scale On any note takcn as a kcy-notc, drivation wc <o?n'c, a d!'M<omtc now proceed to exor to obtain tlie and

its une, are called a /M'7/io?!c scctle.

multiples the whole

of that scries

of a given constitutes

As

is

whoso may bc foundcd, If th key-note, whatevr plain. may bc its absolute pitch, be called Do, thc fifth above or dominant is Sol, and th fifth helow orsuhdominantisFa. TIie common cliord on any note is produced hy combining it with its major and fifth, giving th third, 5 ratios of frequency or 4 5 6. Now if wo take th 1 common chord on titc tonic, on thc dominant, and on the subdominant, and transpose thcm whcn neccssary into the octave

16.]

NOTATION.

lying

immediately arranged Do 1, Hcro the 5 1 T~ 0 with notes tlie Re 9 8'

above in order Mi 5 4'

th

tonic, Sol 3 2' on Do

wo

obtain are La. 5 3'

notes

whose

fre-

quencies

of magnitude Fa 4 3'

SI la 8' 9

Do 2. with with thc

common 3 th

cbord chord

is

Do-Mi-Sol,

ratios

on Sol is SolSiRe, thc tlie Thc chord c 101' on Fa

th ratios 0,

0 still thcsc voice,

~2x~=l:and T samc as ratios. and our bebw above

is i 18 Fa-LaDo, aaby rcpeating

is completed at intervals of octaves. the lower

scale

If we take th diatonic c

Do, or key-note, scale will be d e f p.s to follows th g th a mode

c of a tcnor

h of

c'.

diffcrs Usage slight~y iu wllat different octaves; TIic octave below hoitz. capital letters-C,

I adopt th one just referred

the distinguishing notation of Helmto is written with

tliat with a sufHx D, <&c.; th next below &c. C,, D,, &c.; and th onc beyond that with a double su~xC, side acceuts dnote On th other lvation c", by an octavec', of a violin are written in this &.c. The notes of thc four strings notation, 17. bcen gd~a'e'\ With respect to The iniddie c of th standard pianoforte of pitch confrence recommended. is c'. therc bas

an absoluto At th

no uniform

c' = 2G4 complte to a.' = 440. corresponds Handc!l's or time the 2", ail th c's of 2. This pitch pitch would

practice. vibrations

per Tlie French

Stuttgard second was

in 183-1', Tilis

was inuch have

makes a' = 4-35. In pitch lower. If e' were taken at 256 represented by powers and acoustical note is

frequencies

is usually adopted by physicists and t)as th advantagc of simplicity. instrument makers, ft!) tMt~o of the frequency of a given Thc dtermination

The simplest somo care. method in prinan opration requiring of th Siren, which is driven at such a rate as to ciple is by means In nnison 'with th given onc. Th number of turns givo a note dise in one second hythe la and out which can be thrown of time. interval of a mcasured cncctcd cn'ective othcr holes gives th required ofgreater methods is given by a counting apparatus, of gear at th bcginning and end This multiplied of by th number frotuency. accuracy Th must consideration be deferred. of

admitting

10

INTRODUCTION,

f~g.

So long as we scale of c, th notes above keep to th diatonic written are ail that are in a musical required But it composition. is frequentiy desired to change th key-note. Under thse circumstances a singer with a good natural car, accustomed to perform witliout takes accompanimcnt, an entirely fresh condeparture, a new diatonic structing scale on th new In tbis way key-note. after a few changes of key, tho scale will be quite departed original from.and an immense of notes he used. varicty On an instrument with fixed notes like tho piano and organ such a multiplication is and some impracticahle, is necessary compromise in order to allow th same note to perform different functions. This is not th to discuss the question place at any length, wc will thcrefore take as an illustration th simplest, as wcn as th commonest casemodulation into th key of th dominant. th diatonic By donation, scale of c consists of th common chords foundcd on c, g and f. Jn like manner th scale of g consists of tlie chords founded on g, d and c. Th chords of c and g arc then commn' to th two but th third sca!cs; and fifth of d introdnce new notes. Th thu-d of d written has a t f# frcquency J 3 5 4a removed from any note in th scale of c. 8 4 32 But little th th from interval fifth of d, with a frequc.cy a, whose frcqucncy betwecn the two, two is In ordinary represented note. by by a suitable keyed differs instruments and called a but

18.

is ncglectcd c~ and th or ~?~-Hwc~ M-e identined. 19. Various

compromise

been used th and tliat simplest or at least aimed at, is th equal temprament. On referring to the table of for frequencies tlie diatonic sealc, it will be secn that the intervals from Do to Re from Re to Mi, from Fa to Sol, from Sol to La, and from La to Si, are nearly th same, being rcpresented or while tjj by intervals ~1~' proximate from ?~ relations as exact, Mi to Fa and from equal Si to Do, represented treats into by are

of tomperament systems now most used, generally

have

la

~mperament the octave dividing

~cs~'ap. twelve eqnal

19.]

EQUAL

TEMPERAMENT.

11

parts scalc

called

mean

semitones.

From

thse

twelvo

notes

th diatonic

to any key may be selected to tho folbelonging according rule. the key-note as the first, fill up the series Taking lowing with th third, fifth, sixth, twelfth and thirteenth tcnth, eighth, In this way ail dKScultIes of modulation .notes, counting upwards. arc avoided, as th twolve notes serve as weU for one key as for But this advantagc of true inanothcr. is obtained at a sacrifice Th equal tonation. third, temprament being th third part of is rcprescnted an octave, ~2 :1, or approximately by th ratio th true third is 1-25. than th truc by th interval higher fifth may be obtained from tempered ficmitoncs makc a fifth, wliile twelve ].'2a99, wliile thcrforc ]ow in th cn'or much forte. organ, 20. 2 1, which ratio 1'4!)83 = 1-4.983. The 126 th third is thus tempercd 125. The ratio of th consideration that Th seven ratio is too

go to an octave. The tempered fifth

is thus

is msignificaut;
conse<~uence in

But th

whcn

th

consonance foltowlng

881 883. This 1-5, or approximately and even th error of th third is not of like the pianomusic on instruments quick notes and arc /teM, as in th harmonium of chords is materially impaired.

the twelve notes of the chroTable, giving to th system matic scale according of equal temprament, will be for reference'. is a' = 440 in convenient Th standard employed order to adapt th Table it is only to any other absolute pitch, necessary to multiply throughout
C c

The

by th

proper
c"

constant.

C,
0 C~ D D~ E F F~ G 0~ A A~ B 10-35 17-32 18-35 19-41 20-GO 21-82 24-50 25-95 27-50 29-13 30-86

0,
32-70 34-G5 3G-71 38-89 41-20 43-G5 49-00
51-91

c~

c""

C5-41 C9-30 73-43 77-79 82-41 87-31 92-50 98-00 103-8 110-0 11G-5 123-5

l30'8 138'6 14G-8 155-6 1G4-8 174-G 185' 19G-0 207'G 220-0 333-1 346-9

261-7 277'2 293-7 311-2 329-7 349-2 370-0 392-0 415-3 440-0 4CG-2 493-9

5233 544'4 587"i G23'3 G59'3 C98'5 740'0 784-0 830-C 880-0 933'3 9877

104G-6 1108-8 1174-8 1244-G 1318-G 1397-0 1480-0 15G8'0 1GG1-2 17CO-0 1864'G 1975-5

2093-2 2217-7 2349-G' 2480-3 2G37'3 2794-0 29GO-1 313G-0 3322-5 3520-0 3729-2 3951-0

23-12 L) 4G-25

55-00 58-27 61-73

Zammiuor,

Die J/tMf'~

tx!

<!<c MtMtA'<t<t<cyfc?t DutnottCHte.

Giessen,

18CS.

INTRODUCTION.

[20.

Th gtvcn

ratios

of tho (Zaunuluer)

intcrvals

of the

equal

bclow Note. c c# d d# o f

teinpra.ment

scale

are

Froquoucy. =1-00000 2~'=I-0594G 2 ~=1-122-1 3'~ 1-18921 G

Noto. f~

Froqnonoy. Il 2~'=1-41421
7

g
#

2~=1-25992 2~=1-3348.1 c' = 2-000

L

2'~= 1-58740 Il 2~=1-68179 100 2~~=1-78180 11 2~~=1-88775

2'~ =1-49831 Ii

21. question,
truc within

Rcturning we will
wide

for a moment to thc pbysical aspect what wc shall assume, af'terwards prove whcu two lim its,that, or more sources of th
In th

now

of t!ie to at bc sound any

sum

agitate
point ni

th
the

air

sunultaneousiy,
air, or

resulting
car-passage,

disturbance
is th simple

external

(ni the extendeJ gcomotncal would be caused scuse) of what by cach source Lot us consider acd~g- separately. the disturbanco duc to a simultancous of a note and sounding any or ail of its ]iarmouis. th eompiex By durmition, wholo forms a note having t)ic same pcriod (and thcrefore as its pitch) element. gravcst Wc0 Iiavo at present no criterion hy which th two can bc distmguishcd or thc prsence of th highcr harmonies recognised. And'yetin th case, at any rate, where th sounds have ail indecomponcnt is usually pendent origin-it car, so as to cnect an analysis to say tliat a strictly periodic tion which is not
simple,

difncult to of the mixture. vibration may

not

hy th This is as much as give risc to a sensa-

detect

them

susceptible In offurthcranalysis of point fact, it Ims lon~ been hnowu to musicians that under certain circumstancus the harmonies cf a note may Le heard along w.t!t it, uven w!~n thc note is due to a such as a single source, but tl.e sig.lincancc vibrato strier, of th fact was not undcrstood. Since attention ]~as bccn <1rawn to the subject, it bas becn proved of Ohm and (.nainly by th labours that almost Hchnho~) a)t musical notes are higtdy compound, in fact of thc consisting notes of a harmonie scale, from which in particular cases onc or more members may be missing. Th rcason of the unccrtainty and di~culty of th aualysis will bc touchod upon prcsontiy

but

22.]

NOTES

AND

TONES.

13

22. c:i))ed
writcrs

That hv
on

kind

of note

Hehnhoitz

Acoustics

which th car cannot furthcr resolve is in Ccrmn.n a ')!o?t.' Tyndall and other recent have adoptcd as an Enghsh 'tone' quivalent,

which will bc followed in th prsent Thc work. practice that a. convenient word is almost a matter thing is so important, ~<~ thcn are in gnral of nccessity. made up of tones, th a pitch 23. of the note lu being that of th graves! tone which it contains. m the of dis-

th quality of pitch must bc attachecl to simple toncs only; otherwise th diflcult.y ih'st instance before referred to presents itself. Tlie slightcst continuity in th nature of a note in the effect
experiment

strictness

was oxcmplined th say that hules octave altcr in that Jowcr tho

may lower case of th th slight was

to

change as its pitch by a wholo octave, We should now rathcr Sircn. displacement of thu
ncw

of

alternate tone an to

Introduce

a,

fceble of tlic or

than

pcriod remains vcry nearly In most musical prsent Tho whole. is

any of th

previousiy wholej but as before. howcvcr,

present. th great thc

This mass

is surHcIent

souud

tone th

notes, in sunicient eect of

fundamental

to impress intensity th harmonie overtones

gravent its cliamctcr on is then to

or c/t(M'ac<er 1 of th note, iudcpendently of piteli. modify thc ~ua~~ exists is wcll known. Th notes of a violin, Tliat such a distinction voice with its dincrent vowel sounds, fork, or of th hufnan tuning and yet differ indepcndent~y of &c., may aU hve th sanie pitch and though is due to accompanya part of this ditl'erellce ioudness; to thcir nature as notes, still there ing noises, which are cxtraneous fur. Musical notes may is a part winch is not thus to be accounted as variable in threc ways This we thus be classified First, ~t'<c/t. considered. have already c/tHrf(c<e)', dependsumcicutly Secondly, in which the harmonie ovcrtones are coming on the proportions and thirdly,~oMc~eM. Tins lias to bc bined with the fundamcntal: of comparing taken last, because th car is not capable ('with any precision) character. measure tlie loudness of two indeed in notes which differ much give in onc We shall a future chapter in pitch or a mechanical

of sound, ail of th intensity including system but tins is nothingto th point. of pitch; We are hre gradations of. th sensation of sound, not with a witli th intensity concerned dinerence of loudness of its physical cause. Th mcasure is, as one of more or less wc at once recognised so that howcvcr, Gcrnmn, 'Klaugfarbo'
Moso by Evcrett.

Frcnch, 'timbre.'

Tho word 'chfu-Mter'

iHnscd iH t!)is

14

INTRODUCTION,

f'23.

hardly any choice ~a.rt'&M~ on the magnitude 24;. vibration Wu Luve seoi which

have

but

to

of th that

it as dpendent regard vibrations concerned.

ccc~?'~

a musical

of a secondfor tlie ticking instance, of fi. clock-would not result in a musical note, be th repetition ever so perfect. In such a case we may say tliat th fundamentai tone lies outside of hcaring, the.limits and some of th although harmonie overtoues would fall within them, thse would not ~ive riso to a musical note or ovcn to a chord, but to a noisy mass of likc that produced sound by striking tbe twelve simultaneousiy notes of thc chromatic scale. The experiment may be jnadc witit th Siren tho holes quite by distributing round the irregularly circumferenco of a circle, and turning tho dise with a moJcrato velocity. curs after 25. of By tho construction each complote revolution, The principal is to explain remaining why notes tho instrument, everything re-

cannot evident, of a noise at intervals

is necessarily be truc without

note, as such, is due to a but thc converse, it is pcriodic limitation. A periodic repetitioM

in tlie theory of notes dimculty and tones, are sometimes analysed by thc ear into toncs, and sometimes not. If a note is reallv comulcx is not the fact why immediately and certainly and t)te perccived, of th harmonie compononts disentang!ed ? The feebleness overtones is not th reason, for, as ~ve shall sec at a later staf-c of our inquiry, titcy are often of surprisiug loudness, an(.1 play a promiucnt t On th other hand, if a note is sometimes part in music. perccived as a wholo, why does not this happen always ? Thse questions hve been considered carefully with a tolcrabiy by Hcimboitz', result. The difHculty, such as it is, is not satisfactory to peculiar but may be paralleled in tlie cognate Acoustics, science of Pitysiological Optics. Th knowledgo of external which wo derivo from th things indications of our sensos, is for th most part thc result of inference. an object When is beforc nerves us, certain in our rctin arc and certain sensations excited, arc which wo are produced, accustomcd to associate with th objcct, and we forthwith infer its In the case of an unknown presence. object th process is much the samc. We interpret th sensations to which we arc subjcct so as to form a pretty good idea of their exciting cause. From th dincrcnt views sliglitly reccived perspective by titc two cycs we infer, oftcn claboratc th actual by a liglily process, relief and ~<'m~;t(!)ty)yctf, 3rj oditioH, p. 98.

25.]
distance ~np. lusson of th Thcse in object, inferences
Tbu

ANALYSIS

0F

NOTES.

15

to which are madc

'it~l& life

we

have had no might otherwise with extrme rapidity a.~d quite

bacii ono of

UitCunsciousiy.

ui'

us

is a continued

tho signa presented to us, and in drawing intcrpreting conclusions as to the actualitics outside. Ouly so far as we succeed in doing of any use to us in th tins, arc our sensations ordinary affairs of hfe. TI)is being so, it is no wonder that the study of our sensations phenomena, observation. themselves as falls into th they are called, As an instance of this, it is suNdeiil to mention the. 'blifid on thc retina, which spot' have been might a ~'K))-~ to manifest itself as a conspicuous expectcd phenomenon, though as a fact prohahly-not one person in a hundred million would nnd it out for themselvcs. Th application of these i-emar'ks to thc in hand is tolerably obvious. In tho daily use of our ears question our object is to disentangle from the whole mass of sound that us, thc may rfach parts c&mlng from sources which may interest us at th moment. 'Whcn we listen to th conversation of a friend, wc fix our attention on th sound from him and procecding cndcavour to grasp that as a whole, while wc ignore, as far as them as an interruption. possible, any other sounds, regarding Therc arc usually sufilcient indications to assist us in making this Whcn a man spcaks, th whoJe sound of his partial analysis. voice rises and falls together, and wc have no dirnculty in recogits uoity. It would bc no but on thc eontrary nlsiug avantage, a grcat source of confusion, if we werc to carry the analysis furthcr, and rsolve thc whole mass of sound into its component prsent tones. as regards a resolution into toncs A] though, sensation, position and th practico us to stop tho at thc analysis point, beyond it which would ccase to bc of service in deciphering our sensaas sigus of extcrnal tions, considcrcd objccts\ But it may sometimes that however much wc may liappcn. wish to form ajudgment, th materials for doing so arc absolutely When a note and its octave close together are sounding wanting. and with cnahic single of each there uniformity, perfect whctiicr us.to distinguish, In thc mixture origin. stop is nothing th notes in our sensations a double to or a have might be expectcd, of our lives lead tho necessities of our background,andthat becomc exceedingly subjective difficult of

keyadmits Mat prohubty

part

of tlie organ, the pressing down th wind to a group of pipes, giving a note and tho powor of nttonding to tho inipt-tant nnd ignoring tho
seusationa is to ft great oxtont iuhcntodtQ how great air

nnimportant

of our

oxtont wo shftt) perha.pa Bovcr know.

1C its first sound

notfj

INTRODUCTION. three together,

n!though

[25. pipes
a

or four and
.h~'H

harmonies. th
<ujt

Tho

of each pei-ceived

result
pt'oecuj

is usually
fron

group aiways as a single

cingle

sourco.

is n. of very din'crent with dimculty A considrable effort of attention is rcquired,particu!a~yt).t first; and, until a h~bit bas been formcd, somc cxtcrn:d aid in the slia.pc of a. suggestion of what is to bc Jistoned for, is very dsirable. Th difliculty is altogethcr to that of vcry similar Icarning to draw. From tlic macitinery of vision it might have hcen expectcd that would bc easicr than to nothing make, ou a plane surface, a of reprsentation solid objccts; but exprience surrounding shows that much practicc is gencrally requircd. We s!ia)I rcturn to the question of tlie analysis of notes at a later stage, after we hve treated of th vibrations of strings, with th aid of which it is bcst but a very instructive elucidated duc to Ohm and improved expcnment, originaHy by Helmholtx, bc hre. Helmitohz' may toolc two bottles givcn of th sliapc in the figure, onc about reprcsented twice as )argc as thc other. ihcsewcrc strcams ofair dirccted blownby acro.ss th moutti an<t issuing from gutta-pcrd)a. tubes, whosc ends had been softcnud and prcsscd flat, so as to rcducc thc bore to the form of a narrow in conncction slit, th tubes bchig with th samc bellows. in wn-ter when th note is too By pouring low and by pa.rtin.Hy tlie mouth whcn matter
obstructmg

26.

Th

resolution

of n. note

into

its

toncs componcnt diffrent individuals.

th bottJcs may bo made to exact interval of an octave, give such as b and b'. fhe larger bottic, blown a!onc, gives a somcwhat munlod sound similar in character to tlie vowcl U; but, when bot]i bottles are blown, th character of thc resulting sound is sharpcr, rathci- th vowel 0. For a short time rcsemb)ing after th notes had bcen heM-d Hchnhoitz was able to distinguish separately them in thc mixture; but as the of thcir scparatc mcmory impressions ff)dcd,<thc note scemod to amaJgamatc Itighcr by degrecs with thc lowcr, which at th same time bccamo budcr and acquired a sharper charactcr. This of the two notes bicnding may takc cvcn whcn th t)igh note is th louder. place 27. SeGing now sort that caUcd notes toncs or~y a particular contpound, arc nicapabic of further
p, tf);).

thc

note

is too high, notes with thc

are

usuaDy

and

that

analysis,

7'r'?~M)~/?))~t)yf);,

I 27.] PENDULOUS VIBRATIONS. 17

we are led to inquirc what is thc physical characteristic of tones, to winch they owe their sort of periodic vibrapecuHarity ? What tton a. it, whicii tone ? According to wha.t prod~ces simple matl)cmatical function of t)ic time does tlie in pressure vary th passage of thc car ? No cluestion in Acoustics can be more important. The simpicst with which mathcmaticians arc periodic functions are the circular acquainted functions, expressed by a sine or indecd t!)cre are no otJiers cosine; at aU approaclung them ia .simphcity. aud a<tmitt!ng TIiey of no may bc of any penod, other variation secm well adaptcd to producc (except magnitude), toncs. Morcovcr it lias been proved simple tha.t tho by Fouricr, most gnerai function can bo rcsolvcd singic-vit.hicd into pcnodic a sories of circular functions, Laving periods winch arc submu!tipies ')f that of tho givcn function. it is a consquence of thc Again, of vibration that the particular guttural thcory type, now suggcstcd as corrcsponding to a simple is t!te omy one capabjc tone, of its intcgrity th vicissitudes pt-cscrving which it may among Itave to undcrgo. kind is iiabic to a sort of Any othcr physieat ono analysis, an'ected from anothcr. part being di~crontly If th the car procceded on a dinercnt analysis within from that principle cnucted to thc laws of dead mattor according outside the car, tho consequence would Le that a sound mi~ht originally simple becomo compound on its way to th observer. Thcrc is no i-caMn to suppose that anything of this sort actually When it happons. is added thataccording to ail th ideas we can form on the subject, t)tc analysis within t!tc car must takc place by means of a physical to tlie same laws as prcvail machinery, subject it will bo outside, scen tliat a strong case has Lccn madc out for tones as rega.rding duc to vibrations exprcsscd functions. We arc not by circular to thc guidance of gnera! eutirely considrations like thse. lu tho chapter on th vibration of strings, we shall sec that in many cases theory informs us beforehand of the nature of the vibration executcd by a string, and in particular wliether any vibration is a. component specined or not. simple Hre we have a dcisive test. It is found hy experiment that, whcncvcr accord Ing to thcory any simple vibration is prsent, th corresponding tone can bc hcard, tho simple but, whcnever vibration is absent, thcn the tonc cannot be heard. \Ve arc thercforc justined in asscrtinnthat simple toncs and vibrations of a circular type are indissoluh)y conncctcd. This law was discovcrcd by Ohm. n. ttowevcr left

CHAPTER

II.

IIARMONIC

MOTIONS.

28. time are and

TllE

vibrations

expressed as

by

a circular

function

of

the

variously so important in

designated
Acoustics

simple, ~w~t~M~OM~ or /mr?)M)n'c, thatwc cannot do botter thaii (levote

a cha.pter tu thcir consideration, Lefore cntcring on tlic dynamical of our subject. Thc quantity, whose constitutcs variation part th tlie displacement of a particle mcasured 'vibration,' ma-ybc in a given th pressure at a fixed point in a iluid direction, mdium, and Bu on. In any case denoting it by M, wo have

in

value of u; r is MHp~<(i~, or extreme the periodic after th lapso of which th values <M~e, or jperto~, of u recur; and e dtermines th phase of thc vibration at th from which t is measured. moment which a dnotes of harmonie vibrations Any number of ~e same ~j<M~ affectinto anothcr of th same type, ing a variable quantity, compound wliose clements arc dctcrmined as follows

tho

=rcos(~-A.(2), if i).ud tau ?'=(($acose)'+(SHsin6)~(3), = 2 (t siu e~M ces e.(4).

38. j

COMPOSITION.

19

so tliat s:ud
sounds

if K'=~, t'~cr/b-e,
may

~f vanishcs.
but the

In tliis
expression

case
is

th vibrations
misleading.

arc often Two

to

rather

vcry

mere silence; superposition rest is the consequence, (whcthcr or not) cannot bc so properly called. At Icast if tbis bc Iiitei-furence, it is difficult to say what non-intcrforenco can bc. It will appcar in th course of this work that whcn vibrations exccetl a, certain no intensity tucy longer more compound properly by more addition; bc called Interfrence, nature from that <AM mutual but with it is a which action might pbenomcnon we are now of

cause

propeny but th

bc said

to interfre,

when thcytogethcr of two vibrations

of a totally dcaling. Again, a pcriod,

diiTorent

if tho phases dner cos (e e') = 0, and

by a quartor

or by tbree-quarters

may be reprosented from a pole, tlie lengths of tlio lincs being proby linos drawn to tho amplitudes, and tlie inclinations portional to tlie phases oi' th vibrations. Tbc rsultant of any number of harmonie vibrations is then represented rsultant of by the geomutrlcn.1 th corresponding Unes. For cxample, if they arc disposcd round thc synuuctricaHy of the Unes, or polo, tlie rsultant is zro. vibrations, pcriod 2!). portional Iiarmonic If to we tlie mcasure off an axis of x along takc u for an ordinale, distances we obtain 2-2 protlic

Harmonie

vibrations

~=~"+~. of given

curve,

and timc, or curvc of sincs~

20

HARMONIC

MOTIONS

[29.

whcre quantities

sponding is tlius curvc form motion, direction. ofn. tuning

in is written wavc-!cngt]i, tho range of tlic indcpendcnt dcnoting to a complte rcurrence of thc fonction. the thc and th locus to next of a, point subject a ha-rmonic vibration chapter we shall sec at in

called

place of r, both varia.bic correThe once a to harmonie a uni-

In fork

tha.t

perpcndicuta.r the vibration

so that if an excited is simple harmonie; tuning fork is movcd with uniform to th lino of its velocity parallcl to th end of onc of its prongs handio, fL tracing point attached dcscribes a harmonie in a permanent curve, which ma.ybc obtained fonn by allowing on a piece of the tracing to bcar gently point smokcd In Fig. 2 the continuons linos arc two harmonie paper. curves of thc same wavc-lcngth a,nd amplitude, but of diSercnt

th dotted curve haf thcir rcsu~tant, phases represents bcing <he locus of points tlie two bctween those in which midway curves are met by any ordinate. 30. If two harmonie vibrations of diffrent periods cocxist,

harmonie as a simple reprosented motion with oti~cr clments. If r and r' bc inccmmcnRurabIc, tho value of ?t never if r and T be in th ratio of two recurs but, who!c numbers, M recurs after the lapse of a. time equa.1 to tbo least common of T and r'; but tbe is not vibration multiple simph harmonie. together, period For sounding twicc the exampic, tho vibration whcn recurs a note after and its fifth eqnat are to a time

Th

rsultant

cannot

here

be

of tho graver.

30.

OF

NEARLY

EQUAL

PERIOD.

21 1.

of harmonie of the composition vibrations of dinereut is worth when the dinerenco discussion, periods special na.me!y, ci' the periods Ii' we nx our attention is small. on the course of thiugs an interval of time a fcw during including mcrcly One case
poriods,

wc

sec

that

the

two

vibrations

were absolutely periods equa!, as wc know, to another bc cquiva!cnt harmonie vibration simple 01 tho samc For a fcw periods thcu tho rsultant poriod. mution is approximatcly but tho samc harharmonie, simple monie will not continue to rcprescnt it for long. Th vibration th stiorter on its having period continuaDy gains icilow th dittcrcncc of phase on which th lments thm'cby altering of th rsultant For of statement let us dpend. simplicity suppose that tho two quencies rcpresentcd that when first obsorvod thuir of that vibration othcr; trahze that and cach
abuve

if their

t!ie same as nearly in whic]t case they would,

are

components by ??~ and their th and

cn'ccts of the

conspire,

freoqual amplitudes, ??t?!. is small, and ?!, wlicre At this moment pitases agre. rsultant ha.s an amplitude double a time period (M~) to rclatively 12 thc th

Iiave

But after components. ?~ will hve gaincd ha)f a thc two, boing other. After now in

named,

tion, and rsultant wiLh an that M-M. cqu~ty

corrcspnnding

complte motion

ncudisagreemcnt, a furtiicr of time to equal Mt will hve gained a who!e vibraaltogether aceordancc is once more rc-establishod. T!)e is thcrcfore approximately simple harmonie, comptete intcrval from zero to twicc varying of thse altrations being f<;rks with 500 amI 501 bc frequcnelcs is every second a risc and faU of sound
or bats. opposition of their vibrations.

not amplitude of thc componcuts, If two excited,

tu

but constant, thc frcqnency

tuniug tho'e
t)m is

coincidence ca))ed

Tins th

phcnontenon

We

dn not

hbre

question )icar]y

how etjual

t)tc

ear

behaves

in th

fre'~K-ncie.s, butit of th car almost ccascs for a considrable nelg)ibonr!)ood fractiu)! of a second, thc sound must appcar to fall. For rcasons that will afterwards bats are best hcard wl)en th inappear, sounds are simple toncs. Conscutive notes of th tcrfcring having ni th of th shc\v th stoppcd diapason organ phcnomcnon wcii, at least in th lower parts of th scale. A permanent frence of two notes may be obtained two by mounting crgari pipes Ly sido on of
thc

prsence is obvions tiiat

f~dty discuss of vibrations If thc motion

very Inter-

similar
same wiud

construction clicat.

and Th

identical vibrations

stopped sitic pitch uf th two

22

HARMONIC

MOTIONS.

[30.

pipes little wind.

adjust distance If

thcmselvcs nothing

to can

w:dt th two souud by a rigid could bc eut off, th othcr would bc Instautly restored. Or tbo on which silence balance, dpends, may bc upscb by connecting th car with a tube, whose other end lies close to tlie mouth of eue of the pipes. By grt possible, diScrcnce meaus cxactncss. siuce th of t]te of bats Tlie two notes is object numbor of be~ts of th~t th may be to make in in th thc notes. 30 tuned th to bats
unison with

complete be heard, bctwecn

opposition, th except

so

tliat

at

hissing one pip~s

of thc

as

slow

as

a second

circumstanccs and cognised, two vibrations mercly

frcqucnei.os bats so slow indica.te a ?~M:M<0 on

is oqual to th Under favourable mn,y be renote mighb

a.3 onc lower.

seconds

would

highcr Or it

gains only Le dcsited

to ascertain th diiTcl'ence of thc froqucncios of two notes in unison, in which case nothing more is necessary than nearly to count the number of bca,ts. It wili be rcmcmLcred that t)iG Jifcrcuco of frcqncncics docs uot determine tite tM~erua~ bctwccn tlie on th ?'(t<M of frequoncics. df'pcnds th rapidity of th bca,ts given in by two notes ncariy is doubicd, when both arc takcn an exact octave highcr. notes; AnalyticaUy wlicre Mt M = a cos (27r~< is small. cos (27r?~ e') may cos 2?~ aud wc hve M=r whcre tan Thc with functions = c = e cos(2-7rw< 0) .(1), ?~) t + e M) + e'{ .(3). e] (2), ,n. as harmonie slowly varying Th amplitude + a." + 2aat' cos [Spr (~ e) + a' cos (2?! bc writtcn 27r ()? ~) t e }, e'), two tliat T!tU3 unison

Now

ft sin e + a' sin {Spr ('~ a cos + (t COS{27T (~ rsultant vibration may tLua r and which arc not

~) t + ) 1 bc considcred constant but w M.

clements

of the time, having r is at its maximum when cos {2-7r (?n. and at its minimum whcn cos {2-n- (w thc corrosponding values

th frequency ?~ t + '

e} = + 1, e} == a

n)

e' a + a' and

1, <t' respectively.

beiDg

31.] 31. vibrations that thc Anothcr case

FOURIER'S

THEOREM.

23

corresponding most gericml

of grt is the composition of importance to a tone and its harmonies. It is known function can

bc expressed

nuito single-valued periodic harmonicsby a sries of simple

as Fourier's. Analytical usually quotcd proofs will be aud Thomsou and Tait's fouud in Todhuuter's J~~e~ra~ Calculus and a line of argument almost if not quite ~~M)Y~ r/~7oso~/ty will bo given in this work. to a dmonstration later amounting a theorem arc ail tliat will bo required bore. A fcw remarks A vague is not obvious. Fourier's thoorem notion is not unof arbitrary constants in tho sries tlie innnitudc common that of ropresenting an arbicapacity Tha,t tbis is an error will be apparent, function. trary pcriodic tliat the samo argument would wlicn it is observed apply equally, in which case th expanif one term of tbe series were omitted sion would not in general be possible. of necessity endows it witli Another thc orily point functions, worth notice Instead in a series harmonies are not simple of which it is possible to expand of the simple tcrm elementary is that the

one arbitrarily

given.

formed

by

amplitude serve as wcU as tlie

a similar adding and period. It others

one is for

in th

samc that

vident

of half the phase would thse terms

a~t?! so that tho eacli can term be in Fourier's expressed by sories, means and thereforc double the sum of

sries,

of the

elementary

';,t.t:r.v.

24

HARMONIC

MOTIONS

~31.

terms not,

now auggcstcd.

This

is mentioncd

hero,

becausc

wit~' ~thf) m~y expansions, b~in? nc't"aintcd tiio only oncs (tu:;Jitied arc 1by nature functions h~rmonio simpic function. of a periodic in t!ic dcvclopmont to bo thc clements
Thc rcason of th prccmincnt iinport.a.nce of youncr's scries

students, that imagine

in

Acoustics

is thc

mccha.uic:U

onc

rcfcrrcd

to

in

thc

more and to bc cxp~incd fuHy ))cre:U'tcr, ch~pter, are th oniy vibrations harmonie in guncrfd, simple without a vibrating sun'ering systcm through propagatcd
position.

proceding th:).t, namciy, kind titat arc decom-

32. thcorcm, cfncicnts \Vc may

As in if th

other

cases

of of

a thc

similar

possibility may bc determined writc (1) of 31

character, be expansion

c.g. known, simpio

Tay~or's th coprocess.

by

a. comparativcty

Multip)ying ;). complte period

by

ces from <=C

or

sin

and

Intcgratmg

over

to t = T, wc find

indicating

thn.t ~to is th wcaM value

of 1t throughout

the period.

in of u dpends in tho expansion Thc degrec of convcrgency of th function a.nd its derivatives. on thc continuity ~cnerfd of (1) converge diiercutiations scries formcd Thc hy successive and arithnietical k'ss and loss ra.pidty, but still remniM couvergcut, of it, so long as of the diH'erential coefficients reprsentatives thse lutter
arc cvcrywhcrc

finite.

Thus

(T)iomson

and

dcrivativcs to th M'" inclusive up 77), if aM thc tlic sories for u is more convergent innnitc from values, onc with ] ~< for coc(ncic)tts.
();))' nm) ni )' &c

Tait, arc frue than

33.] 33. tlic bion, ticlc whcn oi' two facility ~cur arc th Another

IN

PERPENDICULAR of

DIRECTIONS.

25

~MS which

compotin(led they Icnd

vibrations, themsel-~cs to

intcresting optical U."

from observa-

with wt~i exccutcd pcriods SM~tM with and

harino~c ~e;~e;~tCM~r are whoc pcnod thc wc not uumbcrs. not curve may many oniy

vib~doi~ ~rcc~ons, commensur~bic, Th timcs is motiorL grcatcr re-cutrant. is

more but thcn th~u in

r~cspecialty the complter tliosc If M cf and th v ratio

pcriudic, co.nponents, ho thc

dcscribcd takc

co-oi-dmatcs,

and dimensions an ellipse, whose position in gnral reprcscnting and upon vibrations of th original upon tlie amplitudes dpend ~er If th phases by a quarter of thcir ph~es. tlie dincrcncc and th quation poriod, co3=0, becomes,

In

this th

co-ordinatcs. tudes,

CMC the axes of th th two If furthcr locus (JcgenGmtcs into

of coinci'dc with't~osc ellipse ha.vc ecju~! amplicfjmpMieuts thc cirete..

which uniform
hn-monic

is described circuiM'
motions,

with motion whosc of thc

uniform may bo directions

velocity.

This

shows

how

into two analyscd arc pori~enJicula.r. and

rcctilmca.r

If thc phases iuto dc"'cncrates

E=0, agre, components liucs the coiticident stmight

the

cl!Ipsc "r"j

When path :dways periods.

the

unison

of the

two

vibrations

remains

in but stcady, pcrfeetly is n sli~it th:i.t there happcn ie timt though TI~o consequeucc

is exact, it practicc
ditTo-cnec

thc will bctwocu

cUiptic ahnnst thc

a f~xcJ

eHipse

rcprcscnts

26

IIARMONIC

MOTIONS.

[33.

thc tho

curve

described

with

ellipse! itsc)f 'cr,noB m t,hu

iyradually
jua~uiLudo

of interest supposing SInec cHipse arc 2(, 26. havo tlic thc

to cojisider thc a and b constants, tho extreme is iti all Sterling cHipsc from cases with concident

for a fcw perlods, accuracy in -jon'cspondence with t)io cttangos of e. It becomcs thcroiorc a matter of ellipses system rcprescntcd by (2), but j variable. are in thc in i a, t b respcctivcly, whose sidcs rectangle or 6=0, wc agrcemcnt, = 0 its As of u and

sufHcient

values tlio

insct-ibcd pitascs with ellipse

tlie

emcrcascs
Leon) es

C to ~-n-, thc

dia"'ona.l opcns out until

equation

From other

tins point

it closes

up ngam,

ultimutely

comciding

with

thc

= 0, to thc incrcMc ofe from eon-csponding ~Tr + diagonal to 7r. Aftcr t!iis, as e mngcs from vr to 2~ th dHpsc retraces Its course untU it again coincidcs with t!ie first diagonal. TIio of changes is exhihitcd in Fig. 3. sequoice

Thc dctcrmiucd

ellipse,

having by its point

a,lrca.dy four given tangents, of contact P (Fig. 4) with

is compictcly thc linc ~=&.

33.] ]
In order to connect

LISSAJOTJS'

CYLINDER.

27

this

with and

when

~=6.

of tvvo hn.rmc:nc uf th BUpcrposHion GJIiptic paLhs with the e va.ries uniformiy coincidont of ncarly pitch, vibrations vibration a. l)n.rmo)uc cxccutcs a.lorg ~J.' 7~ itself timc, so that frcuf thc twu givcn to th differenco witit' a fi-cqucney equal qucncics. 34. rcr'-arded d~cnbcd
Lissn.]ous'

cos27r?:<==l; bc tl~ ~tit-

e, it is Ruflcient !t=acos. thercfore

to

observe Now if

tha.t th

bas thc surface

shown

as thc ou

diffrent

of ellipses may be systcm tlic Sfimc enipso of onc and aspects In Fig. 5 of a. transparent cylinder. that this

~Z~T! Seen tangent

thc represents from n.u infiultc at J. to tlie and thc turns

cylinder, distance

of which in th

~1J3' is a plane section. direction of tlie common

plane

rectangle, cylindcr Its own

ellipse upon its romains

is projcctcd into a tlie cylinder sections, now that thc into its diagonal. Suppose section with it. thc plane axis, cai-rying a constant rectangle in which th pro-

projection

the posirepresents jcction It a right angle. through tion of tlic cylindcr in round we obtain tlio cylinder that by turning thereforc uppcars to thc pa-ths described by ail th ellipses corrcsponding succession of equal pcriod and iixcd vibrations to two harmonie a point subjcct be turned if tho continuously Moreovcr cylinder amplitudes. of thc ellipse is inscribcd. a rotation after Fig. 6
1 ~tHM~s de CAtM~ (3) LI, 147.

28

HARMONIC

MOTIONS.

[34.

with wc

uniform obtain a

velocity, complte

which

insurcs

a harmonie of thc uf thc

motion

for

dcscribcd

by thc point differ eacli complte slightiy, loss of a single vibration'. thus
composition

rcprcsuutation wh~n Lhc periods revolution Th bats

varying two compunents

.P, orbit

synchrouous
of

vlt)i thc two

th

to a gain or answoring of th cyliuder arc rvolutions rcsult f)'om thc which woutd wcrc to act in thc s.uuc

vibru-tious,

if they

direction. 35. rcn.Hxt'd iixud

action

Vibrations

of thc

Mnd

hre

considercd pondulum-bob,

arc

very

expcrimeutn.))y. point by a long wirc

of gravity,

A Ii(j:Lvy or string,

which of

may

circumstunc'e.s But that in order

projection, to sec th orhits

descrihes in particular cases, into straight pass to thc best

hung cliipscH undcr according lincs or

easily from a t))c th to

circles.

be described so thcy sliould ou th retina madc by th moving point at any part of its course bas not time tofade heforc tl)e point cornes round again materially, to its action. This condition is fulfilled by th vibration of a silvered att~ched to bead (giving a straight in a vice clamped the lines of light. under th (such as a knitting-necdie), tiie system is set at the lower end. When which appear luminous ellipses, point dcseribcs Thse ellipses influence of would friction contract in gradually until t!iey subsidcd undergoing owing any to somc othcr want by reflection mctaUic wire a luminous point), winch is

advantagc, tl~at quic)dy

it is necessary th 'Itnprcssio!i

firmiy into vibration, as into fine dimensions

a stationary bright wcre it not that change, of thc symmetry, in the which th orbit wire th is lias plane cumstances already 3G. vibrations sitnpiicity

without point, in ail probability, s)ightly vibration sceu to

to according ditiering puriods is cxceutcd. Undcr thse cirt!io cycle of changes undcrgo

cxplaincd. So far we Itavc supposcd tho of th periods thc next case othcr. Wc component in ordcr of have

to be equal, or nearly cqual; is when one is the double of tho M=acos(4~7!-<e),

~=Z'cos2?!7~. of t may bc written

Tlie

locus

resulting

from

thc

limination

Dy

vibration

will

aiwaya

ho

mcaut

iu

this

work

comj)~<<!

oyclo

of

chfUtgOB.

3G.]

CONSONANT

INTERVALS.

29

which angle

a curvc for ~1 values of e reprsenta 2ct, 2&. If e = 0, or 7r, wo ha.vc

inscribed

in

the

rect-

reprsenta iutcrvals

p~bolas. of tLc octave,

FIg. 7 shcws thc aud fth. twuifth,

various

curves

for

U~c

by of represontation method Lissajous' the relative and whcn is applicable, th transparent cylinder of circ~mstanccs diffrent from th whctber is altcrcd, phase dor continuously owing to a sbght in diiferent cases, projection will th cylinder in tho ratio of tbe poriods, from exMtness viatior. of th to the eye digrent aspects so as to prsent to turn, app~r sa.DiO line traced on its surface. To aU thse systems a vibrating in arranging system so dinicutty vibrations shall consist of two harmonie of a point that th motion ratio. in any assigued with their periods in perpendicular planes, A wire as Blackhurn's known pendnlum. is that The simplest at tbe samc Icvel. two nxcd at ~1 and points ~t C-B is fastcncd wirc CP. to its middle by another point Tbe bob P is attached thc point of suspension ofthe diagram, in th plane For vibrations stretched that th wires are sunIcicQtIy C, provided iH practically 37. There is no

30

IIARMONIC

MOTIONS.

J37.

but

for a motion tho

D, can-ying

to this plane, th porpendiculinwire ~O'j9 witli it. TIic pc.ri~s

bob

turns

about in

of vibration

thc

principal DP. Thus octave.

~pproxnnatc

planes

arc in the

if ~C=36'~ To obtain tito

unison,

Y~

rdativeJy 3S.

small.

of t!.c square roots of CPand the bob describc.s th figures of thc of curvc.s squence to correspondin~ must bc so tiiat is ncarly tight,

ratio

Another contriv~nco called thc was originkalcidophonc ally invented by Whcatstoiie. A straight tllin bar of steel carry'i~ a bcad at its crid is fastcncd in uppcr as vice, in a cxpMncd If the section of th bar is previous p~ragraph. square, or circuleth poriod of vibration is indepeudcnt of thc plane in which it is But let us suppose pcrformcd. that the section is a rectale with sidcs. Tlie unequal stress of tl.c bar-tho force with which it rcsists thcu grcater in t!te plane of Lcndin~-is mc.ater aud tlie Huc~nc.ss, vibrations in this phuie have th shortcr pcriod By a suitable of tho thickncsses, adjustmcnt the two poriods of vibration bc into any required may brought aud th eorratio, curve cx]iibitd. responding Thc defeet in this is that thc samc arrangement bar will r.Ivc one set of figures. only In ordcr to ovurcome tins objection th fullowlng modification lias bccn deviscd. A slip of steci is takcn whosc section is very rectangular so tliat as ciongated, in regards onc plane the stiHhcss bcnding is so gr~t as to amount to rigidity. Thc bar is divided practically into two parts, and the

38.]

OPTIOAL

METIIODS.

31

broken throush

ends

reunited,

the

two

a rigtit angle, so that ~<j.,tt'.inK thc LitickucfiM oi' ojf:, rod is clamped tlie compound tion, th period of th vibration of tho on th Icngth cntircly that point 39. thc the in t)]C second th direction lowcr at which In this

oa one another pices bcing turned tho plane, which contains th small
gi'L'ut thujkMt~ i' tho ti~i-. W:

at a point bolow the juncin one direction, dpend ing alinost but pice, is nearly constant; uppcr in a vice by varying th

may be controlled pice is clamped. th luminous

vibrations

arrangement which are to bc

the point rcaiy fixed, while its experimont, into motion of successive apparent by means M~Mf/e is thrown A smaU hole in an opaque mirrors. from two vibrating reflection of light, scrcen placed close to the iiame of a lamp giycs a point in th mirrors reneetion of a after which is observed by means of polished steel, arc attMhcd usually forks, and th whole is so disposed to th prongs of stout tuning into vibration th luminous that wlieu th forks are thrown point motions in pcrpendicuhn' harmonie to describe directions, appears of the renccting motions surfaces. Th to tho angular owing small tlescope. The mirrors, motions of these harmonie and periods dpend upon amplitudes thoso of tho corrcspnnding forks, and may bo made sucli as to give witli tlic witli cnhanced brill.ianey any of th figures possible it is possible to project arrangement kalcidophonc. By a similar In cither case they gradually contra.ct as tho ri~ures on a scrcen. the vibrations 40. of the forks die away.

obscrvcd of light remains

itself point but in Lissajous'

excutes form of

Itavc reccived an important of this cliapter Th principles of rectilinear motions. in the investigation periodic application of a sounding is a particio Whcn a point, fur instance string, as to give a note within thc limits of with such a period vibratiug is much too rapid to be followed its motion by tl~e cyc hearing, to know tlie character of th vibration, so that, if it be required Th must be adopted. thcomcthod somo indirect simplest, retically, uniform as when in a perpcndicuhu' direction, Instead curve on smoked a tuning fork dra-ws a harmonie paper. use of a revoltlio vibrating of moving body itself, we may make an M~K~e in motion. In tins us with ving mirror, w!iich provides of tlic function charactcristic of a. reprsentation way we obtain to timc. with thc abscissa tiLe vibration, proportional th is to compound motion of translation vibration undcr examination with a

33 But it often

UARMONIC that happons or inconvnient.

n'<u a.~tDu~ui~

MOTIONS, of this we
'fbU~i'bL'

[40. mcthod substituts

))nri.).1 in

the

ho dimcult
thc uniform

application Jn such cases

vibnt.t'ofi i

would for
i.h<'

may

that th point, suppose whose motion to invcstigatc, we wish vibratos with a vertically th result of combining period T, and let us examine witli ttus a horizontal harmonie whose is somc mu]tip)o of 7-, motion, period say, M/r. Take a rectangutar and with axes parallcl pice of paper, to itsedgcsdraw th curve th vertical motion rcprescnting (hy off abscissa3 to th timc) on such a scale that sctting proportional tLc papcr jnst contains ?~ repctitions or waves, and then bend tlic with a re-entrant paper round so as to form a cylinder, curve runA point dcscribing this curve in sucli a manno' ning round it. ideas, that appear punod it revolves about uniformly from a distance to combine T, with a horizontal harmonie to obtain th th axis of given motion of pcriod tho reprsentative t]ic will cylinder vertical motion of ~T. Conof curve th

same

direction.

To

fix our

let

us

in order versely thcrofore, tho vertical vibrations, the

must

bc imagincd to he into a piano. Thcre developcd and th situation uf thc curve upon it, \vitcn thc adjustcylmdcr ment of tho periods is not for thon tLe cylinder exact, quite to turn, and the contrary motions serve to distinguisb appears those parts of th curve which lie on its nearer aud further face. 41. Th

cylinder dividcd

apparent path a gencrating Une, and along is less difnculty iu couceiviug thc
containing

harmonie motion is generally obtained auxiliary of an instrument called a vibration-microsc-opc optically, by means invented One prong of a large tuning fork carries by LIssajoua. a lens, whose axis is perpendicular to th direction of vibration and which may be used cithcr of by itself, or aa t!tc object-glass a compound formed of an eye-pieco microscope by tho addition In either case a stationnry is independently supported. point thrown into apparent harmonie motion a lino parallcl to along that of tho fork's vibration. The and it will vibration-microscope of the law universality bc found may be appHcd to and test th rigour TIms

a pure musical when examincd strict unison with frequeucies

connecting pitch that any point of a vibrating note will appear to describe witb a vibration-microscope its own. of By th same the consonant

~ep't'o~. body -\v)uc!) gives a rc-entrant curve, \\hosc note is in means thc ratios may of be intervals

characteristic

41.]

INTERMITTENT

ILLUMINATION.

33

verified;
acoustical

for this latter though mthode to be described

a more purpose thoroughly in a future chapter, may be

prcfcncd. 42. Another method of examining use of intermittent of suitable thc motion illumination. apparatus intcrvals T.
a

of a vibrating Suppose,
series of

body dpends upon thc for exampic, that by mcans cleetric are obtained sparks body,
must

whose appear

If,

Itowcvcr, little, the iHuminatcd to vibrato of the thon whose The

at regnfar A vibrating of thc sparks period is also T, cxamined by thc light at l'est, because it can be sccn only in one position. of th vibration differ from T cvcr so th period and the varies, position a frequcncy which is thc of the body. Th type from body will diffcrcncc of an vibration appear of that can

slowly ~ith and tliat spark with sries

be observed

facility. of sparks can

bc obtained

Induction-coih

circuit is periodicauy broken fork, primnry by a vibrating or by somc othcr intcrruptcr of snrRcient But a bette)' regularity. rendered intermittent with tlie aid of rcsult is afforcled by sunlight a fork, whosc of meta], parallel to prongs carry two small plates the plane of vibration and close togethcr. In each plate is a slit of th fork, and so placed as to aAbrd a pM'aIIcl to thc prongs fj'cc passage throug)i th plates whcn th fork is at rcst, or passing th middte of its vibrations. On th opening so through point is concentrated a beam ofsunHght by means of a burningundcr is placed in th cne of examination glass, and thc object on thc furthcr When tlic fork is made sidc'. to rays diverging vibrato by an cicetro-magnetic thc illumination is eut arrangement, off exccpt when the fork is passing Us position of equithrough formed,
librium,

or nearly not
so

so. with

The thc

nashcs
as clectric

of light
sparks

obtained

by this but

method when a in my bc takcn is thon

arc jar

instn.nta.nouus

is connected

th rcguhu'ity exprience to eut on' extrancous ]ight very striking. A similar result

wire sccondary is more perfect. as

far as possible,

(especially of th coil), Carc

and

shoultl thu cnect

at by looking at th vibrating may bc arrived a sries of holes arranged in a circlc on a-revolving hody through (tisc. Several sries of holes on the same ma.y be providcd is not satisfactory without <tisc, but th observation some provision for sceuring uniform rotation. Ti~ier, 2'/ti/Vn~. H. Jtm. 1807. 3

34

HARMONIC

MOTIONS.

[43.

with respect to the sharpness Except the samf when the pcriod of th light
th vibmtin~ ~c'y. Tiiis pouit. ~HHt

of definition, is any multiple

bu att,ciided

the

result

is of.
th

of tt~t
tu ~i)eu

revolving When that

wheel the

is used frequency

to determine of intermittence

an unknown is an without Titis

frequency.. exact multiple apparent condition of th motion, of things vibrations

of

t!te object is seen vibration, but generally in more than one position. is sometimes advautageous. Similar and for as of th example, grt effects flashes th arisc number when ratio of th are in th

of th

frcquencies of two smaU in

whole

numbers.

vibrations

stationary,

as the number again and in general double.

of flashes,

a given time be th body will appear

If, half

CHAPTER

Iir.

SYSTEMS IIAVING ONE DEGREE 0F FREEDOM.

43. concerned,

THE matcrial usually of very varions are

and are suscomplication, of vibration, modes any or a!l of which ccptible Indeed in some of th moment. at any particular may cocxist most important aa strings and organ-pipes, musical instruments, is theoretically and th number of independent modes infinite, the consideration of several of tliem is essential to the most pracof tho consonant chords. to the nature relating is often in which one mode Cases, however, thcmselvcs, present of paramount and cvcn if this were not so, it would importance still be proper th consideration of thc general to commence proquestions blem need with not th simplest be supposed case-that that th of one mode of frcedom. degrce of is th only treated It one occur tical

with systems, of considrable

whosc

vibrations

Acoustics

is

possible, because their possibility 44,

of other modes do not so long as vibrations is of no moment. under other circumstances

one degree of frecof a system possessing dom is denncd M, whose origin by th value of a single co-ordinate to thc position of cquilibrium. TIie may be taken to correspond Mnetic and potential ofthc for any given position system nergies TIte condition arc proportional respectively r=~~ whcre selves w and to tlie to and F=~(i), of M. But if we Hmit our-

functions are in general consideration of positions

M!. the ?'y~:e~'<~e u is a small this

?:eK/Aquantity, wo

&~u)7iOOfZ o/</Mt< con'M~on~t')~ e~x~t~, and m and are sensibly constant. On

understanding 3-2

36 now

ONE

DEGREEOF

FREEDOM.

[4~. from internai

without,

If there he no forces, cither rcaulting proceed. from or imprcss'~d on the systcm friction or viscosity, Thus remains constant. \vhole energy y+ Substituting to tho respect for T and time, 1~= constant. V their

the

wc obtain

and differentiating tlie e~ua-tion of motion values,

with

~m + /tW = 0 of which th complte integral is a) vibration. by th It nature

(2)

~=(tcos(?)< whcrc scc ?~=/7):, that thc

(3), will bo

a ~Muc rcprcscnti))~ alone is detemuned pcriod the

of the

itself; system cumstances. say, if T werc

any restriction,

on cothttcral cirand phnse dpend amplitude wcrc exact, that is to If tlie difrercutial equation and F to to thon, without strictly proportional th would vibrations of th system harmonie. bc accuratc)y
ahont its conDguration

of equilibrium of cases tlic assumption depends approximation not to push limits.

But

is only approximate, propoi'tionaHty ?< is always that tlie displacemeut nature of the and required th application

majority on an dcpending smallhow small

in th

on th

and tlie degree of system particular be careful thon of course we must its proper of th intgral beyond

the prinwithout a limitation, not to be stated But, although of a configuration of a system about th vibrations eipic that of th on th structure have a period cquilibrium dcpending of tlie vibration, circumstances and not on the particular system is of suprme or th practical givcn importance, If sidc. whcthcr thc pitch wcre regarded and th not from loudness within wide th theoretical of th limits note insuch

by a musical thc art depcndcnt, as th violin Th

Instrument of th

a.nd pianofortc, time periodic

on many instruments, pcrformer would bc revolutionized.

so that

an increase

in w, or a decrease

of a vibration.

By a generalization th case of a matcrial particle urged brium by a spring, ?~ may be called th

thc Juration in /t, protracts in of the kuguage employed towards a position of eqnHiinertia of th system, and

44.]

DISSIPATIVE

FORCES.

37

of Thus an augmentation spring. u. th force of th quivalent th perK'dic t.imc. of incrcas<?s By mass, or a rc!f).xation spring, limits for wc may somctimes obtain means of this principlc which cannot, or cannot the value of a, period, easily, he calculated cxact)y. 415. idal Th absence never of atl forces but of a frictioual character is an

to in practice. only approximatcd is aiways sooner or Tho original dissipated cnergy of a vibration is another source of loss, into leat. But there latcr by conversion case, reahzcd which dissipative, yet produces speaking, properly the case of a tuningConsider results of much thc same nature. friction will in time stop in ~fMMO. TIic internai fork vibrating into will bc transformed and th original th motion, energy
though not,

heat.

But

In space. a single tute In attempting, prohicm, cultics, sirable. wc and

suppose tlie strietness

now

that fork

th and parts exact

fork the cannot

is transferred air surrounding be ti'catcd

to au open it constiseparately.

whose system, tlie Ilowcver, sliould gencrally in

solution

bc stopped

a of so complicated diniby mathematical

solution would be deany case an approximate a few periods is quite insigThc crfect of thc air during We are and hecomes nincant, only by accumulation. important of the motion which its effect as a ~s~<r~?:ce tbus led to considcr Ttie disturbing force is periodic take place t'~ ~acKO. would (to th so), and may he and one proportional to tite acclration, dividcd Th former produces th same offect as the other to the velocity. in th mass of th fork, and we have nothing more an altcration Th latter is a force arithinetica.Hy to do with it at present. prosame approximation into two parts, to thc velocity, and portional and thcrefore motion, produccs In many those duc to friction. communication aiways enccts similar acts of cases in thc to opposition same character of motion as that head the as by duo that th vibrations are

th loss

to dissipation sumcicnt tion with a degrce of approximation to th velocity. poses by a tenn proportional 0. M-T XM+ H"M ==is tlie quation subject of vibration to frictional M=~e'~

undcr th same may bc trcatcd and is reprosentecl in th proper,

diScrential for acoustical

quapur-

Thus (1) one is dcgreo of

frcedom

for a system with The solution forces. <(}.(2).

cos (~i~.

38 If thc friction be

ONE

DEGREE that

0F

FREEDOM.

[45.

thc > solution its changes and no lorger fonn, to nn os<-Hiatnry motion; f'orrRsp.nds but In .di acousticai A: is a small applications 'Under Dicso qu~ntit'y. circumstances as cxprcssing (2) mny bc r~u'ded a harmonie vibratton, mett-ical whosc Mnpiitudc progrc.SHio]), is wlicn not constant, consi~o-cd but aft-cr dimiuishcs cqu~l successive iutcrv~Is cxtronu m gcoof

so grt

time.
excursions

Thc
Is

difercncc
nc:u-)y

of
consent,

th

t]tc Logar:t!imlc Ducremfut. It is cxpresscd if T bc thu puriodie by ~r, timc. Titc on ?~- ~~Invo!vG.s frcquotcy.dcpcnding on]y tite sccotd to thc rir.st order of approximation powcr of A:; so that ~e/c~'o~ /t(M ?!0 e~ec~ o~ ~c y)en'o~a principe ofvo-y gnera! appiicatiun. Tho vibra~on iicrc consided is ca!]ed thc/y-ce vibration. It is tbat cxccutcd hy thc System, when disturbcd from cquiHbrium, and tbcn to itself. 4G.
Jcss ibrcc

logarithms :md is ca)!ed

of

Wc must

now turn bchaviour

oui- attoition ofthc

funetion of

to anothcr whan
timc. In

problem, subjuctud
ordcr case in tu ijicludinnoui-

not to a
savc

Important,th varying as a

systcm,
thc

harmonie

wc may takc at once the more gcncral rcpctition, friction. If tho-c be no friction, wc bave on)y tu put /< = 0. Th dincrential is quation

rcsults

This

is caDed

a. /M-c<~

to a force coutinued

Imposbd opcratioa

vibration; upon it from wititout, of that force. T]ic

[' it is thc

of thc responsc and in mainta.iued

System by tho

amplitude

is proportional

46.]
to ~thc as that Let system produced magnitude force. ofthe us now of in

FORCED VIBRATIONS.
of tlie force, and the period tLe th effuut is the on

39
same

supp<jHu gi~uu, ahd a variation in tlie period dinfcrent cases are not

trace of

force.

a given The effects

of th vibrations frequency produced which we are t)ie force, and thcrefore variable in th comparison th cncrgy of the about to institute. Wc n~ay, however, compare in different of passing the cases at th moment through system thus to specify th moment of equilibrium. It is necessary position at which total one the of energy part energy is not the is to be computcd invariablo througitout in each tlie the case, because vibration. During

hecause tlie similar; strictly the samo as that of is always

force, impressed back again. From (4),

from the tho systcm reoives energy period and during th remainder of th period yields it

if u = 0, cncrgy ce when ce shi~c, suie==l, or, from (5), wc bave by p=n. If

and

is thcrefore

th maximum

a maximum, kinetic energy

bc denoted T=~sm~(6).

The when

kinctic the

encrgy

of the motion

is therefore

vibrato internai and (5),

of the force is tliat in period the influence of its fruciy undcr forces), ~t0i<t ~h'c~'o?! Th

the grcatest possible, which th system would own

ela-sticity (or othcr vibration is then by (4)

and,

if

quarter Thc Since natural

bo small, of a period

is very grcat. amplitude bohind that of tlie force. = ?!, may tlio actual also be treated vibration

its

Its

phase

is &

case, where tho period of to thc system,

IndGpendentIy. is the same as that

40

ONE

DEGREE

0F

FREEDOM.

[46.

If p bc Jcss tha.n ?;, the rctardation of phase to tho relatively force lies betwech xeru and a qu:u-te!is ~reater pcriod, aud whcn tit:m}.[.,butwcchi(.~U!u'~t'~(.'i:m!n.i,.bntfuut~d. In t!)c cusc of a systcln devoid of i'riction, tlie solution is

ttian ?~ thc pl.ase uf tiiu vibration When is amaller agres with tliat of thc force, but the sign of th vibrawhcn~ Is th grever, tion is clianged. Th change of phase from complte agreement to complote which is graduai disagrcemeut, wlien friction acts, hre take~ place abruptty t!ic value 7t. At th as pa.sses through samc tune thc expression for th amplitude bccomes inanit. Of course this oniy means that, iu thc case of friction cqual periods, 7~<~ he taken into account, Ijowever smali it may be, aud liowevcr its rcsult wben insigniricaht and ?t are not approximatc!y cqua). Thc limitation as to th magnitude of thc vibration, to which we are all along subject, must a)so bc borne in mind. in one direction th generating force is at in tho opposite as happons, for cxampic.in du-eetion, the canal theory oft!ic tiftc.s, is somcti.ncs considcred a paradox. that Any dimculty may be fc)t will bu ronovcd the extrme by considering case, in which th so t!.at thc natural ".spring vanishes, is Innnitety period In lono-. fact we nced ody consider the force acting on the bob of a'common pendutum swinging frecly. in which case t]ic excursion on one sicle is greatest w)tcn the action of gravity is at its maximum m thc opposite direction. When on thc other hand the inertia of th system is very sma)I, we hve the otticr extrme case in which th so-c.Ued bccomes equiHbrium theory tlie force and applicable, excursjou being in tlie samc phase. Wi~en t]te pcrioJ of thc force is than the nature longer period, thc cncet of an increasing friction is to introduee a retardation in th ph:Lsc oft)tc zero up to nquarter dispiacementvaryingfrom penod. the period If, ),owever, of th natural vibration bc tho longer, th original retardation of haf a period is diminished by short ofa quarter somethmg or th cn'eet of friction period; is to Mc~e tlie phase of thc cstimatcd disphccment from that eon-cto thc absence of friction. spond.ng In cither case th influence of fr.ct~on i to cause an to thc state of approximation things that wou!d prcva)I tffrictioTi wcre paramount.
at That th excursion shouid bc

whi!e

its maximum its maximum

46.] If vnry a force

PRINCIPLE of

0F

SUPERPOSITION. with slowly untd th free

41 vibrations thc disforce lias

s1o\v1y

nearly equal period to a maximum and then not rcach its maximum

pjacement

docs

dcrte, aftcr th

Under thc opration of the force at its to diminish. bcgun thc vibration continues to increaso until a certain limit maximum, this incrcase is approachcd,and continues for a time cven att))ouglt tlie ttds ucw force, and 47. motion
forces is

its having passed tlie t'utardation principic full From rcsulting

thc simple

maximum, of spring cxp]ained'. of the 1.

Ou begin.s to diminis)). tidcs bchiud tlie da.ys of

luoou lias bccn tlie linearity from thc

sum

it follows that the cquations simuItanGOus action of any numbcr of of tlie motions duc to the forces ta~en tlie or in vibration absence a manner proper to of any othos. transmitted into itself, Thc th

force Each scparate!y. to tlie wthout regard of a force peculia-rities

c:uises

presoicc arc thus For

motion

if thc force be periodic in example, timc T, so Each ])armonic elevibraLion. resulting ment of tlie force will call forth a corresponding harmonie vibration in tl system. But since tlie rctardation of phase e, and the ratio is not the samc of amplitudes M for th different components, system. will be th the vibration, resulting though periodic in c/t<t7'KC<c?' from the force. It frent that frce one of
thc components is isocbronons,

of tho

in th may

same

happcn, or ncurly

time, is diffor instance, so, wit)i th

in whicli case it will mauifcst itself in thc motion vibration, to its original out of al] proportion As another importance. the case of a System actcd we may consider on by two example Th resulting forces of nearly cqual period. vibration, bcing comin unison, is intermittent, of two ncarly to the pounded accordiug in thc last chapter. pt'inciples cxphuned To the motions, which arc tlie Immdiate effects of t])c imbe added thc tcrm expressing frec forces, must pressed always if it be desired to obtain the most gencral solution. vibrations, Thus in thc case of one impressed force,

48.

Thc

distinction

betweenybrce~and~'ee

'vlbra.tioQS

is very

~Airy'B2'(~t'<n))~n'at'f~Art.328.

42

ONE

DEGREE OF FREHDOM.
bu olearly understood. by thc force whicli
~hHu Lii:t ut' thc

[48.
Thc pcrioJ of t))c to act
un)y

and must important, former is detenniucd

.u Um h~ lic-ni

solcly
~rdiQut:,

is supposed
htttcr dpends

itself. Anothcr system point of din'ercnce is that so long as the extcrnal influence continues to opcratc, a forced vibration is permanent, being rcpresentcd strictly by a harmnic buta frec vibration function; begraduallydies away, aftcr a timo. fur cxample, coming ncghgibic that the Suppose, systcni is :),t rcst when thc force 7~ cos ~j{ bcgins to operate. Su.ch rinitc vaincs must bc givcn to th constants jd and a iti (1) of 47, that and ii arc initiatty buth zro. At first tllen tiiere is a frec vibration not less important than its rival, but after a time it to insignificanee, and the forced vibration is left ill complte of the nc!d. Tins condition of things possession will continue so long as the force oprtes. Wlien thc force is removed, thcrc no is, of course, in the valucs of M or !<, but discontimuty tho forced is at once convcrtcd vibration into a frce vibration and the poriod of thc force is cxchangcd for that natural to the system. thc coexistence of the two vibrations lu thc earlier Dm'ing part of thc motion, tho curious of beats phc'nomcnon may occu)', in case the two periods diiicr but siight)y. For, ?! and being nearly conditions smali, tlie initial arc approximately equa), and satisfied by !< = a cos (~< Thcrc sensible. stages is thus
TI)is

on the constitution

ofthe

friction

rednces

e)

e'

cos ~1motion,
vcry

ej. so long as e' in the remains earlier ( G3).

a risc

and faU in the

Is

intermittence

conspicuous

of thc motion

of forks

driven

by cicctro-magnetism

49. two to ways distinction

Vibrating according ofpitch more

of one degree Systems of the to t)tc values is sumcIe)tt!yIntG!!igibic; closcly Titc the

of freedom constants but

may vary in M and K. Th it is worth w]nle

of a grcatcr or less con'sefjncnccs most obvions is the more or less dcgree of damping. rapid extinction of a fi-ce vibration. The enbct in this direction may be mcasurcd of vibrations wliich by the numbcr must e)apsc bcfore the amp)Itudc is reduced in a given ratio. Initit )y tho amplitude may be takcn as unity; after a time <, lot it be 0. Then 6 = c'

examine

40.]

VARIOUS

DEGREES 2

OF

DAMPING.

43

If =
nnjucru-Lu

~T, wc have a; =
du~reu ut' dampmg,

xT log
\vc tua.y

la

a, system
tn.kc

subject

to

on)y a

upprox.UTmtcly,

Thi.s gives

thc

number of

of vibrations

which

arc

performed,

heforc

thea.)nplituduf!iih)to0. Thc inuucncc

is aiso powcrfu~y Mt in a, forccd damping wlicu thcre is a. :uear approach to isochronism. In the 'Ibra.tion, case ci' an exact equality a.nd ?~ it is thc damping alone betwcen thc prcvcnts thatwheu auticipate dcvia-tion from witich motion tUc
becommg

m~nite. is small,
wuld

We

might

casily

damping

isoein'onism perfect in thc Magnitude of thc vibration, thc s:uuc precision of adjustmcut

a. compara.tively slight cause a large fa.Hmg off with not damping, bc rcquired. From a larger

but that would

tlie

enuatious

bc sm~l!, must bc very nearly equa.1 to 7)j lu. ordcr to a, motion ]iot grea.tly Icss than th maximum. producc Th two principal eScets of damping may be compared by betwecu Th result is climijiating (1) and (2). so that

if

where the the

th

sign

of tlie

square

root

must tlie

be so cliosen

as to

make in

right-hand If, when ratio

sidc ngative. a system vibrtes after x vibrations of the resulting

frcely,

in th will (p), thc energy of case of perfect isochronism in the ratio T T~. It is a mattcr mdiifcreucc whcthcr th forced or tlie free vibration bc th higher; all dpends on the M/erua~. thc intcrval In most cases of interest tlie formula p = ~+8~ may be written, is small; and then, putting

then, motion

ampUtude wben it is acted

be reduced

on by bc less than

a force

44 The givcn

ONE table following by Hcimlioltx'

DEGREE c~culatcd

0F

FREEDOM. from thcso formulte haa

['49. been

Ijttcrvfd
of Uto

con-cspon.Ung to a rduction
ruM.ttauco y y to ouc-touth. ~Q i'itcnHity

v'Lmtif.nM of a frco

nfter vibrntiou

whiuh is

tho ro-

ducudtoono.tunth. ~=A.

tonc.

~.oo'~ 19'00 9-50

? Whuif! tuno. y tono 7 toile. Twu whuit' tonea~ major third. = minor thu-d. tonc.

G-33 4-75 3.~0 g. 2-71 ~'37 y j

Formula

(4)

shcws

that,

w!ien

i.s small,

it

varies

c~~M

'1. Mtt)~~ 50.

as

a: observations of forced vibrations of a system due may to known

From

forces, tlie natural period and tniGd. TIio formuhu are

dampiug

Le deter-

If tudes

tlie equilibrium tells us tlie value

thcory of

Le known, sav

tlie

comparison

of ampli-

~ifH~/ntJf~fyc~

p. 221.

50.] ] and e is also kuown,

STRING whence

WITH

LOAD.

45

51. frcc th forced

As bas

been

ah-cady

vibra.tions

is important which vibrations tbcir

a system, system, vibration to

take which may

origin and thc first, influences thus have to be reekoned limits are fixed in dctcrmining a wider view of thc
in. question will

of forccd and the distinction stated, that most of but it may be remarked as affecting we shall Lave to consider of a second in the motion ultimately is innuenccd in cvcn of the embracing as forced A by it. its relation when force as free. both that which th An

a system whose lias a share pystem

acts Systems, upon thc it.

arbitrarity, thc period matter

be

On
vibration

recognizcd

example is part

in air A tuning-fork tliis clca.rer. vibrating ma.y ma~c th air and itself, and of a compound including system But is free. the vibration in respect of this compound system of th air, yet thc thc fork is influenccd by thc raction although it is For practical is smaU. of such innuence purposes amount and Lhat of fork as givcn, if th f<c<:ta~ motion No crror will be committed th air as forced. be takcn as tbe of the fork (as innucnccd by its sun'oundings) of tlils mode of But th peculiar basis of eaicutation. adva.ntagc solution in thc case of an approximatc is manifcstcd conception for th actual It may then sumce to suhstitute bcing rcquired. of of th fork in the absence would bc tbc motion what motion, if uecessary. a correction, introduce air, and afterwards
convenieat to eonsidcr

the

motion

of the

52. drawn

Illustrations from

applications simplicity

of this chapter may bc principles Wo will give bc're a few of Acoustics. ait parts of their on accouut an early deserve which place of the

or importance.

two nxed points bctwccn A string or wire J.CJ3 is stretched to carries a mass J~ which is supposed and at its centre ~1 and itself ncglith mass of the string bc so considrable as to rcndcr of equilibrium, asidc from its position is pulled WIten gibic. whicb th lino C~ vibrations, and thcn Ict go, it excutes along C'.V= x. Th tension C= 6'~ = M. are the subject of inquiry. on the amount of equIHhrium of thc string in the position dpends In any othcr it has been subjected. to which of the stretchiug

4G the position of vibrations fraction treated

ONE tension

DEGREE

0F

FREEDOM.

[52.

is ~reatcr but we limit ourscivcs to the case so small that tt~e additiona! is a ncgJigibJe strotching of the who)c. On th~ (~ncHii~n thc ~<i)i rn:~ bc as con&tant. We dnote it by y

Thus,

Idnetic

cncrgy

= ~;

amplitude cumstances, being conccrned. Equation

Th

and

phase

arbitrary

of course on th initial dpend so far as th dinforcntial quation

ciris

th nanner in which 7- varies with each (2) expresses ofthe Independent quantities V.~a: resultswhich bc may all outained of the (~MCHs~~ by considration (in the tcchnica! sensc) of the quanti tics involved. T!~G argument from dimensions is so often of importance hi Acoustics tliat it may bc wcll to consider this first instance at Icngtit. In the first wc must assure place ourselvcs tliat of all th on which quandties T may dpend, th a only oues involving

53.J

METHOD

0F

DIMENSIONS.

47

rfrence mnssure wnLLeu-

to a,

thc

three and

fundamental T. Let th

unitsof solution

time, length, of the problem

and bo

This l)e the arc

equation fundamcntal

must units

rcta.in

its form

nnmerically that in deriving nitudes dpends, sions arc arc of those T is the
(Mass)

by means as is vident, expressed, it no assumptions would units. oniy

(Length)

unchanged, of which th when. be made it

whatever four is

may

quantities considered magwhich f

as to th

on quantities and since its dimentime onc involving it follows tl~at whe!i ? and ~f (Ti.me)'

New

of all

tlie

otherwise a change in th unit of time constant, oc.T' would necessarily the equation disturb Tins being (3). admittcd, it is ca~y to see that in order that (3) may be independent of the unit of Icngth, we must Imve r ce T"~ Is constant and n~, when finally, in order to secure indcpcndence of the unit of mass,

Therc docs not

must

be no mistake Iiave

as to what nMKMte~

this

on no other periodic climenquantities, having sions in spacc, time, and mass, t!ia.n thosc aLove mcntioncd. For we hve not proved that r is indpendant example, of thc amplitude of vibration. so far as it is truc at ail, is a consThat, cf thc linearity ofthe quence dinercntial approximate quation. From the neccssity of a complte cnumeration of all the on which th required quantifies rcsult thc method may dpend, of dimensions is somewhat but when used with proper dangerous Ct~re it is unqucstionably of great power and value.

Wc prove. time dcpeuding

that

argument thcrc is

does and a deHnitc

ONE

L.MCREE

0F

FREHDOM.

F~.

1 lie solution of th prsent problem bo made thc might foun~tion of a ,nethcd for Lhe absolute n~asurerncnt of pitch. pnncip~J to accuracy impedunc-nt would t!to prubabjy.be oi difBculty u~ku~ iu relation to th m~ suf!ictcutfy i~ of tlie ~u.c, without at tlie samc timc th note too much in lo~crin~ th musical scalc.

.'3

T)ic fur hcr

wirc

may

bo

strctc)ied bndgo

en<) beyo~i wouidbcc:dcu);),tcdfrom

wcight or pulfey at

by

~t~chcd Th pcnodic

te

its

time

T).c in fect,

ratio aud~=

of ~.2.

= ~i, tl~c pcriodic musical

t).e balance. If r. be ,no..sured timc is exprcs.sed in seconds. the Instead of being over its !cn~ of th n~tu.'c oith two t)te same

~n~ concentratcdin an

~evertiK.Icss the gr~vest more c osoJy, supposingthc c~cs at th jmddie point.

.vcight, th centre, is uniformtydistnbuted t)ic prsent problem gives some ide~ vibration of snch Let t.. string. amplitudes

compare

of vibration

-When tLro~h

th th

uniform position

is .straight, string of its cquilibruua,

th dirent

moment

of passin~ are a~ parts

54.]

COMPARISON

WITH

UNIFORM

STRING.

49

mated thc

with

a. variable

centre.

velocity, increasing If we attribute to th whole th at kinetic the

stretched

from mass

either tho

end

towards

that centre, it la vident ovcr-estimated. Again,

th uniform string i.s more

cnergy ci' moment

than

will
maximum its

of tho vclocity bG cousidcrab!y

excursion, winch

suhstitutc,

foltows tial "i

thc

cncrgy th mass

courses ~1~ straight is dumnished y t)to at the middie a;= 0, and according timo. For

and substitution.

the poteuaccordingly TIiG concentration tho when kinctic ~-= 0,

whcn cncrgy and thercforc, the

at once increascs point decreascs ttte potential energy to the principle explained m

pcriodic catcuiatud from that ~1/ dnotes t)jat in order

a string thon the period the formula of the last section, on the supposition th mass of the string. It will afterwards appear to obtain a correct result we should !)avc to takc in4 TT is duc Of 4 thefactor-~ TT hy far thc more import-

44, prolon'~ is less than that

stcadof.Von!y-~V. ant part, 55. viz.

tu (,he difTcrcnce

of tlie kinetic

nergies. but one thc

of a System possessing example practicu.Hy one dcgree of freeclom, let us considcr tlie vibration of a spring, end of which is clamped in a vice or otherwise held fast, wliile otiter carries a heavy mass. this !iko tho last lias strictncss, System an innnite numbcr of Indcpcndent modes of viwhcn thc mass of t!tc spi'mg bration is but, vibration which is sn-i:d), ttiat )-e!ativc!y ncarly
indcpcndont most important of its t!)at inci'tla. tho buconics so much th

As another

In

othcrs

may

bo

Pusinng

this idca,

spring merety attaehed mass and, if a certain

to it.s limit, we may as tite origin of a force urging th towards th position of equilibrium, point be not excecded, Thc in simple result is a

ignored. the regard

to thc disp!acement. proportion harmonie with a period on vibration, dpendent th stinhess of tho spring and the mass of the toad. 56. In

of tho oscillation oi' the centre of inertia, consquence H~ci-e is a, constant towards the communication of motion tendency to tlie supports, to resist which th latter must be a.dequate!y In ordcr very ni'm and massive. to obviate this inconvenience, R. 4

~0 two

ONE

DEGREE

0F

FREEDOM. bc mountcd

[5G. ou

cqual amp)itude in such a, manner that the motions arc a.Iwn.ys opposite, or, {m it may otherwise bc e.xprcsscd, with a phasc-tiiHcrcucc of !m]f a period, thc centre of inertia of thc whole system rcmains at rcst, and thcro is no to set thc fra.mc-work into tendency vibra.tion. We shaU sec in a future that chapter this peculiar relation of phases will quiddy estab)ish itself, wt~tever disturbmay be tho original a.nce. In fact, any part of tho motion winch does not conform to the condition of Icaving thc centre of inertia unmoved is soon extinguished by dampindccd thc supports ing, unless of tbe system arc more than usually nrm. wc found a rough illustration exemple of tho fundamental vibration of a musical so hre with tlie string, and attachcd load wc may compare spring a uniform slip, or bar, of elastic one end of which material, is securejy such for fastencd, instance as the ~:<e of a )~e~ instrument. It is truc of course that tlie mass is not coucentmtcd at onc end, but distnbutcd over th whole on account of tlie Icngth; yet smallness of thc motion ncar the point of support, th inertia of that part ofthe bar is of but little account. ~e infer that thc fundamental vibration of a uniform rod cannot be very dincrcnt in cbaractcr from that which we ])ave bcen Of course for purconsidering. poses rcquiring the two Systems arc sufnciently prcise calculation, but where t!ie object is to form clear idcas, distinct precision may often be for simplicity. advantagcously cxchanged In the same we may regard spirit tlie combination of two and loads shcwn in Fig. 13 as a reprsentation springs of a fork. This tuning which bas been much instrument, improvcd of late years, is to the acoustical indispensable On investigator. a large scale and for rough it may bc made by wciding purposcs a cross piece on the middle of a bar of steel, so as to form a T, and then bending the bar into t!io of a horse shoe. On th shape handle a acrew should be eut. But for th botter class of tunmg forks it is prfrable to slape th whole out of one piece of stecL A division from one end down the middic of a bar is first running 57. As in our first

similn-r prccisely tlie same frame-work If thc two loads pcrform

and lo~ds springs m~y in a symmctrical manncr.

vibrations of

S~J

TUNJXGFORKS.

5j I

thctwo madc, and th whole shape.

to a plane

ont to form the prongs parts opcned workcd and n!u into by tho iiammer prongs must bc cxactiy symmctricat the axis of thc liandie, of incrtia direction may remain in which

of thc with

the fork, rcquircd

T)ic two
passing

t!ie during unmoved, vibrato. Thc

through vibration th centre tiiat is, in thc

respect in ordcr that

unmoved

thc

prongs

is cnected To make t)ms. tuning th note higher, th incrtia of thc System quivalent must bc rcduccd. This is donc hy nling away t)ie ends of thc prongs, cithor their diminishing or thickncss, thcm. On the other actuaiy sliortening to hand, Jowcr the pitch, tlic substance of the prongs ncar thc bcnd may be rcduced, the effect of which is to diminish th force of the t)te inertia Icaving or the inertia pmctically unchangcd may be increased which would (a mcthod be prfrable for tcmporary pm-poscs) thc ends of thc prongs by loading with wax, or material. forks arc Large somctimus with movcothcr provided able weights, which slide along thc prongs, and can be nxcd in any position As thse approach by screws. thc ends thc (whcro is the quivalent vetoeity greatcst) incrtia of thc incrcascs. System this way a considrable In range of pitch from may bo obtained one fork. TJ)c number of vibrations for any position per second the weights on th prongs. may be markcd of Tite relation bctwcen the pitch and thc''size of tnnin~ forks is In a future simple. it will be provcd chapter rcmarkably that the material provided remains thc samc and tho shape constant' tt.c period of vibration varies, dircctty as t)te linear dimension if t!ic linear TIrns, dimensions of a tuning fork be doubicd, its note falls an octave.
.spring,

of a tuning fork is a ncarly pure tone. Immeafter a fork is struck, diateJy high tones be hcard, may indccd to modes of vibration, whosc nature will bc subsecon-esponding but thse qucnHy considered; die away, and cven whilc rapidiy they do not b!cnd with th propcr s they exist, tone of the fork of thcir part~y on account and partly very high pitch, bccause do not bchng to its harmonie ihey scale. In the forks examincd Helmhoitz the first of thse overtones had a frequcncy ~.by from 5-8 to n-G timcs titat of the tone. proper forks are now Tunmg with rsonance generaUy supplied cases, whosc effect is tu the volume greatly and augment of the pnrity 42

58.

Thc

note

53
sound, according oiJer to excite
')~ :~dr~t.t

ONE DECREE
to thon,
.Cr~~

OF FREEDOM.
to or be ccHo
~'u

[58.
dcve!opcd. wcll supp)icd In with

principles a -viotin
rh< prongs'u

hcreaftcr bow,

dit'<<t)uuof\'b''ai.u)r'.

Thc

souud R~.

so prothccd

wIU last of
varies

n minute

or more. T)~ and with th

As standards
organ-pipcs

pitch
with

pitclt

of

tuningforksarcinvaluabic. tlic temprature

wltio]) cnn with th tension, th~t of strings of t!ic wind; pressure fork kcpt ctc.m but n. tuning for long; constant nuvcr be rctaincd or magnctof temprature to violent and not subjccted changes its pitch with grt fideUty. ixation, prcscrvcs forl. may bc copicd with tuning By means of bcats a. standard is iu a second of bats !)card Th nnmbcr very grt prcision. of thc twu tcncs which produce cf th frc()uencics t))u dinurencc
thcm; a minute so that cach, if thc bcats frequcncics can 1)0 diH'cr madc so s)ow as to occupy hah'

ti)C grcatcr

tion. (~ptic:tl

Still incthod.

precision

might

of a vibrahy on)y l-3()th be obtaincd by Lissajous

of observation, In consc<)ucnc bcing dimcult is duc to ofi in thu sonnd a faHing whcthcr of tho vibrations, or tu thc gradut interfrence dying away Ho took a fork modihcd a sonicwbat plan. Schcib)cr adoptcd or from tho standardwbcther in pitch highcidiffrent ~ightiy tbc countcd but wc will say, tower,and lo~cr is not materia!, fuur About togctbcr. 'Tmmber of bcats, when they were soundcd Very sh)\v bcats uf tho unccrtainLy
bats a second is th most suitab)c, and thse may be countcd for

is then made sligbt]y to bc adjustcd to givc wit)t it prccisdy fork, and tuncd than the auxiuary higbcr lu tins way ft of beats, as did th standard. tlie samc numbcr perbaps a minute. Th fork
copy as exact a~ possible is secured. To facilitate Ute counting

of

thc

bats could T)ie

tbe

Scbcibk'r

cmployed

pendulums,

whose

periods

of

vibration 60. dtermine tuned 'to

bc adjusted. mcthcd
al)so]ute

of bcats

was

to bridge prcparcd tliat cacii fork gave with its so smaU ovc-r thc stcps that could ofbcats hi t! sries a numbcr immdiate ncighbours to of frcqucncy T!tC din'urencc con'csponding be casily couutcd. Thuir sum, with aU possible accuracy. each stcp was observcd of th octave, for the intcrval of fi'cquencies being tlie din'crencc formcd th starting to thc frcqnency was 'quai of that fork which an octave, lotcrval by of others

of pitch and a number

aiso employed lus standards.

to by Scheibler Two forks were

(;0.'j point at couldbc th

SCHEIBLER'S bottom of tho sries.

TONOMETER. Th pitch of th other

53 forks

dcduccd.

C.5 in a.ll will forks givc four bats per second, If conscutive frora c' (2.')G) to c' (5L2). over thc intcrval to bridge bc rnquirod th but it is probably is laborious; mctitod Un thisaceountthc of pitch, as it Is liabtc dctcrmina.tion for tl)C original most accm-atc It will clhninatc. as care and repetitioti to no ct-rors but such
mn.y bc obscrvcd tliat thc cssctYtiat thing is tho mcasurcmcnt of

of frcqucnfor two notes, whosc ~o of frcqucncics ~er~ce If wo could be sure of its accm-a-cy, known. cics is UKlcpcndcnt]y bc suhthc Intci-v:d of th nfth, fourt)i, or cvcn. major third, might number of rcducmgttie for th octave, with th advantagc stitutcd tttat with thc aid It is proba.b!c of th ncccssary interpolations. bc succcs.s(ut!y t))i.s course adoptcd, as of optic!d mcthods might and a.rc casily rccognised, thc ngurcs Lissajous' con'csponding with whicb test of t!ie accm'acy is a vcry sovcrc thcit- stcadinoss tt'e ratio isattainud. forks may bc detcrmincd by of large Thc frcqnency tuning which on smokud curve to trace a harmonie papcr, aHowiug them of a rcvo)ving on thc circumicrenco bc mountcd tnny couvcnicnHy of thnu givcs in a second of wavcs cxccutcd Thu muubci' drmn. t)tc thcfrcqucncy. In many in 4-2 givcs frcqucncy. (il. is vcry )nusic:d A scrics uscf\d of forks for
th

cases

tbc

nsc

of Ittterniittcnt )net))odof

Hturnination

duscribcd

a convcniunt

dctcrmining

an

nnknown

ranging
dcturtnination

at snndi
of

int.crv:us thc

over an octave of any a~o

may and is caUcd Schuibtcr's note, case In cilber a note tu any desirct). pitch. bc nscd for tnuing of beats thu f')-u(Utcncy of th note is dctermincd hy tl)e nuinher to it (on cach lie aearest thc i'orks, which \vhic)i it givcs with sidu) in pitch. For tuning pianofortcs thc notes of or organs, a. set of twelvc forks may be

frcqucncy It Tonomctur.

tho equal tempsealc 0)1 cbromatic thc uscd giving are notes Tbc or any dcsircd corrcsponding rament, system. It is betto-, to unison, and t])C otbcrs tuned hy octaves. adjusted th forks so as to givc four vibrations per I~owevcr, to prpare note is thcn tuncd Eacli k-ss than is above second proposed. tlie corresponding fork, until than thcy givc when little higher It will be in thc second. four bats sounded cxactiy togcthor

54

C~K

DEGREE

0F

FREEDOM.

[61.

ubservcd to thc

that
frcfptencics

tho

addition
is not the

(or subtraction)
samc

of a constant as a more

number

thing

displaccnicut

ofthescatcinabsolutcpitch. In th ordinary tlie other notes that

so thitt on

practicc dctermiued twefve

tho

of tuners

a' is takcn

from
(Iftt)s.

rcmcmbcrcd
uct.ivcs,

of by cstinm.tion truc ~fths arc slightiy tcmpcranicut System

a fork, and It will bc of seven f)ft)~ is :).

succci-isivu

in excess
cn.ch

equal

little fifths, ordor Hfths wurk

fit.
coining

Thc

tuner in

down

procccds upw:irds au octave aftcr about

from every of the

Le

Ly altet'ttate scfdo.
thc

tu reimutt should must

britig

the same part nearly ])itn back to . If this all the

thu samo

stop, m Twcivc
case,

not

the

as nearly biu o-ror sensible Th

bc

bc ruadjustud,unt,iJ as can bcjndgcd, is

thcn hnpartiaUy

twe)vc
sma!)

ftfths

amount.

arc too fhtt by, Thu incvita-

as possible. fo!Iowii]g numbers

as little di.stributed, an<t rotdcrcd Tt)c octaves, of course, arc all taned truc. indicatc thc order in whic)t the ilutes may
JJ y' a' M~ b' c" c~ c"

takcn:

c' c'~

e'

13 51G 81911 314 6 17 9 1 12 415

In

7 18 10 3

thc cqual is only approximatcly atpracticc temprament but this is pcrhaps not of muc)t consquence, taincd cousidering titat the systcm ainied at is itself by no mcatis pcri'uction. other Violins and Instruments of that class arc tuncd by truc nfthsfrom'.

G2.

In

illustration whosc
jnution.

of n. pendutum
xouta! harmotuc

let us consider the case of/o;'C6(Z vibration is subjoct to n small huripoint ci support
is

thcboba.ttachedbya.fincwu'c to a movcn.btc 7~. 07'*= point 7'() = and .r is th honxoltal co-ordiniitc of (). SInco tlie arc supposed vibrations sina.)!, thc vertical and motion tho ~cgiccted, thc wlrc !n:).y tension Le of

Cfjuatcd to thc wcight of (,). Hunce for t))c Itorizonta!

motion ;e+~+.(.t;J=0.

C2."]

COMPOUND

PENDTJLUM.

55

New tlie form

so that p)itting~=)~, oe cos~<; trea-tcd of, viz. ah'cady .v + A:~ + )~ = cos~ is limited
motion

our quation

takes

If~)
Thc mc:ms of

l)e equa.1 to ')!, thc

horizontal

motiou

o!i1y by

the friction.

a~sumed

harmonie

a, second it
in its

pcndulum
motion.

of massive An rings cfncicnt scrowcd

.P witli

rcajized by maybe which can'Ies construction, is shewa in arrangement into a beam, or other nrm

forP

t)tc ngnrc.

/t, .H arc iron

bar, which carrics :tttac!K!d ncar its ends, and is supportcd cqua! hcavy weights A', in a hurizo)it,al at riglit to th beiuri by a wirc position angles thc fuur rittg.s in thc tnanner shcwn. Whcn tlie passing through i.s )nndc to vibratc, n. point m thc rud midway bctwcot pcndulutii support;
rings attachcd

C', D similar

to

a stout

C' and

excutes

a hiu'mouic

motion

ni

a direction

6'D, and jnn.y bo nmdo th dutunt If ttte wcights -Z~. to

of a.tta.chmcnt poitit A~ and be vcry

to paridtcl of auother pcnin relation grcat

Q, t)jc uppur penduhun swings vury ncn.r)y in ils 0~1 propur in () a. furccd vibr:<.tion of titc s!t.nic period. poriod, and induccs \Vhcn thc ]c!)gth ~Q is so adjusted that thc nattu'id pc!'i<j(!s oftite two pcnduimns arc nearly t)ic s;unc, Q will bu tLrown into viuk'ttb evun t!)&u~h thc vibration ot' j! bc of but niconsidura.bln motion, ampHtudc. ln this case the diHcrencc of phase is about n (~)artcr

5G

ONH

DEGREE

OF

FRKEDOM.

[62.

th uppcr pcndulum of a. pcriod, by winch amount thc vibrations hf vnry dHTo'cnt, If the two pcriods in p!)asc, to or arc compictcly opposed accordin~ and (5) 63. n. ibrk of 4C.

is in a.dvancc. ~either quations a.g)'ec (4)

A vo'y good cxa)np)c of {t furccd vihruttcn i~i abrdcd by of tui intermittent thc iunucnec ctcctric under cui'rcnt,

~hoso
sma!)

period
c)ectro-magnct,

is ncarly shape shcwu ~nd supportcd

cqual
formed

to its
by winding

own.

~).CZ?
insula.tcd

is the
wire

fork;
on nn

7?a
iron

corc

of tho

armature'), an intermittent
acts upon

the

ni E (simila.r to titat known betwccu tho prongs ofthc current i.s sent through th wire, a This force is not cxprcssibic fork. by but
functions,

as 'Sioncn.s' fork. Whcu force periodic a simpic cirtheorcm lu a

nny with of If

cular
scrics

fonction;
of sucli

mn-y bc
ha.vlng

cxpandcd
poriods

by Fouricr'.s
T, T, T, isochronous

&c.

thcsc, furk,

of not thc latter

too small will be

insigninca.nt. plete
to its

pcriocl
first

othcrwisc vibrato In wbat follows wc will suppose that T whicb ncarly ngrcc.s witlt tliat of the expressing th pcriodic vibration,

amplitude, canscd to

bc ncarly

the is com-

t]tc

effect

it is the

scqucntly lu cai'cfuHy

rega.rd thc sries

term.

furk, and couforce as reduccd fork must be

order

to obtain

t))C maxitnum siiding

thc

tuncd hy a small

is cfpud friction) pcriod (without is approacticd, Witen tho desired trial. cloue by actual c~uidity from rc'st, and comto start and th fork is a)!owcd thc ibrccd ptctncntary frequencics, produce
y"r txrjK'ntino.

pice orby w:LX', uutit its j~turat to that ot' th force. Dus is bcst

frce and

~ef<
Uu~ j'urposc

of nearly cqual therefore ( 4-8) in thc bcginning is a measuro of whose stowncss -vibration
\\nx mny <'onvp))ifnt]y Lo fioftcncd Ly

arc

amplitudes

a.nd motion of
a )itt)<i

of thc the
)nc'IUnK

accm'acy
it wiU)

G3.]

RELATION

0F

AMPLITUDE

AND

PHASE.

57

lias bad a-fter tl)c froc vibration It is not until adjustmcnt. ch~'acter. assumes its peru.anent time to subside, that th motion and inounted fork properly constructed ofa tuning T))C vibrations a vcry slight to very little arc subject damping; consequcntly off a markcd occasions isoclironism falittig from dviation perfect of the rsonance. in th intcnsity with can bc obsci-ved of thc forccd vibration The nmpHtudo verifibut th exprimenta! suficicnt accuracy by thc car or cyc its phase bctwccn out by thcory of thc relations cation pointed at'rangcand that of th force which ca.uscs it, re<~ures a modined tho
mcnt.

Two cluded
current.

similar in
thc U])dcr thrown

cicctro-magncts
samo thcsc intn circuit,

arc

acting excitcd
It

on

similar same

forka,

and

in-

circumstances sunDar

by the is ctear

Intermittent

tha.t

will bo

Tliis by cqual forces. Lot at)d amplitude, iu pcrpendicula.r effected Lissnjous'mcthods

becausc vibrations, thcy rcfcrs both to phase of vibrations similarity vibrations arc the now that us suppose of of one directions, and by mcans

Systems arc actcd on

th

The resulting arc opticallycomponndcd. ngure from tho case in which thc lino. a. straight is ncccssarity Starting of both viz. whoi tbo natural are a maximum, pcriods o.mpHtudes as that of thc force, lot onc of them bc put a same forks arc tbc their that whatevcr It must bc rcmonbercd out of tnnc. little unison in perfect vibrato natural periods may be, the two forks Tho principa.1 with onc another. with thc force, and thcrcfore is to destroy the of tbc natural Ciffcct of thc dift'urcnce periods
synchrolism of phase. Thc straight hue, which prcviousiy rcprc-

an bccomcs vibration, so long as th forks i-cmaius steady, perfccHy a quartcr thc forks arc botb period Originally ]owcred, it falls ~Vhcn thc pitch of one is slightiy scnted the compound
the force, and at thc samc timc its amplitude

and this ellipse, arc not tonchcd. behind still thc force. bchind
Let titc

more

diminishcs.

diifcrcncc amplitudes

thc two forks bc e', and betwccn of phase Thcu (t: (t.. of vibration by (H) of 4C M = Mycose'.

tlie ratio

of

58
The a.nde'. following e

ONEDEGREE 0F FREEDOM.
table shows th simu1ta,neous values of a

[C3.
c< 0

e 1-0 -!) -8 .7 'C -5 '4. '3 -2 -1 It direction amplitude. othcr may tbat a. considrable 0 2550' 3C 52' 4.T' 3-t' 537' GO" 66"25' 72 32' 78 27' 84. 15" altration of 1

appears

phase

ni

either

without may be obtaincd very Whcn one furk is vibrating be made to dinfcr from it on

materialty at its either sido

th reducin"' thc maximum, by as muc)t as or by as

CO" in phase, without lo.sing moro than t)alf its much a.s -I<5", without losing more tha)i Iiaf its 45" in advance, and tbc ing one fork to vibratc of t)te phase corresponding to t]ic c:~c of maximum obtain a phase diScrcncc of 90" in conjonction amplitudes. G4. Tbc Lissajous' intermittent invented by ngurc then bccomes is best T)tia

amplitudu,

e?M)'~y. By aHowothcr 45" in arrcar

rsonance,

wo of

with

an cquality

a cir~e. obtaincd Ly a fork-

current Hchnbottz. as before.

interrupter and cicctro-magnet

mountcd

arc connected, ono witb ono po!c ofthcbattcry, and th othcr with a mcrcury Thc ot]ier ple of tbc battcry is connectod cup. with a second A U-shapcd rider of insulatcd mcrcury cup. wirc is carried over thc cups, at sucb a by t!)c lower prong just Iieigbt tha.t during the vibration th circuit is altcrnatejy made and brcken of one end into and out of thc by titc passage mercury. T)ie other end may bc kept pcrmancntiy immcrscd. By mcans of t!tc pcriodic force t)tus obtaincd, thc cnuct of friction is comand thc vibrations of th fork pensatcd, maintamed. pcrmancnciy In order to set anotbcr furk into forced vibration, its associatcd
ctcctro-magnct

of a fork may consist TIie wires of thc ma~nct

maybc

includcd,

either
p.

in tbe sanic
li)0.

drivix'Y-circuit

?'<)'))t;~?)t<~o~t't),

G4.]

FODK

INTEBRUPTER.

59

is effected by another whose periodic or m a, second, interruption luto mci'cury rider dipping cups'. ui seti'-act-ing instrument is of tm.s kind Tho ??~(~<& ~/ If the force acting on th fork often imperfcctiy apprehcudcd. whetlier tlic circuit were open on its positionon wouid work donc in pressing any position ttirough aftcr a, complte therc so that on tlie return, bc undono period would be nothing outstanding by wliieh ttie effect of thc frictional whic!i docs not forces could bc compcnsa.tcd. Any explanatiol only dependcd or eloscdtb take mark. accouut Thc is wholly bcside the of thc currcnt arc two and causes of retfM'datiou irregular contact, th touches merWltcn the point of th rider nrst of' thc rctardation

scJf-mduction.

on account of contact is imperfcet, probahly cnry, thc cicctnc tlie mcrcury tho On thc other air. ha.ud, in leaving adhcring of tlie hquid in the cup to is prolonged contact by the adhsion th currcnt is retarded wire. On botli accounts thu amaigama.tcd to thc mcrc position of the furk. behiud wliat would correspond only on the dcpended would still be rctarded by its selfposition of th fork, thc current th contact However hiduction. may be, a finite current perfect the lapse of a finite until aftcr time, efumot bo gencrated any eau be suddenly a finite vclocity mechanics more ttian in ordinary But, evcn if the resistance causes arising", the From whatcvcr on an tuert body. impressed is that more work is ga.iued by thc fork effect of th rctardation tlian is lost of tlie rider from tlie mcrcury, the retreat during remaitis to be set off thus a, balance and durin<T its entrancc, t]tc position of the fork, onlyon depcuded bc considcred to th phase of' its first harmonie component nught Thc reof that of tlie fork's own vibration. be ISO" in advance 1 1 Lnvo arr<mgc<l aoveral iutcrruptora on tho nbovo pJfH), un t)io componont Tho forks woro mado by tho vilittgo blucksmith. n)trtn being of homo manufacture.
t!)C magnetic

of the

circuit

against If

friction. force

Tho

Scmo further entl being ticrowod down ou tho bo.so board of tho instrument. tuoms of adjuating tho IcYcl of tho morciu-y surfaco ia necosMry. lu Hcimholtx' intcrruptor a horso-.shoo cloetro-magnot embraemg tho fork in adoptcd, but I nul inctmod to profur tho prosent arranHcmcnt, nt auy rate if tho pitch bo low. In somo cases a greater motive powor iHobtuinod by n horffo-fihoo magnot acting on n.
Kuft iron Mmftturo carried horizontally by tho uppor prong aud porpoudicuhtr to it.

eupn

conxiat.od

of

iron

thimbloa,

(ioldored

on

ono

omi

uf

copier

slips,

tho

1 h<woususUy found a singlo Smco cull suicieut buttery puwor. Any desired rctardtt.tion might bo obttdued, in dcffmH of thor mcans, by attnching tlio rider, not to tho prong itscJf, but to tho fnrthor oud of n liglit Hirnight spriug cnrrie by tho prong and Bet iuto forccd vibration by tho motion of
its point of nttttclnuent.

60

ONE

DEGREE

OF

FREEDO~f.

[G4.

taxation

ofrcdnccs apoken be rcdueed to 90", th force

~tU.'t.x~t

this advance. acts in

th most

If

th

phasc-diu'crcnce
manner,

favourable

p'-K-~bh.~Yibmuon.i.sptudtiL-cfJ.

It is important to notice t))at (cxccpt in thc case just, rcfcrrcd of ttie mterruptcr dKFcrs to some cxtent from to) the actual pitch tbat natur.'d to thc fork according to th hnv cxprcsscd in (5) of e being in thc prsent case a. prescribcd pbase-difcrenco on t!)c na.turo of thc contacts :ind <Lo jnagnitudc depcnding of thc selt'-uiducti.on. If thc Intermittent currcnt hc empioycd to drive a, second maximum vibmtion i.s ibr]<, thc wlien thc freobiained, of thc fork coincides, not with thc natural, but with tbc 'jucncy modHic-d frcqn<jncy of t)te inten'ttptcr. 4G, Thc dviation of pitch is practica.Hy is possible, is a.t nrst a. tunmg-fork very smitt); but intcrrupter th fact that from such its natur:d a dviation

Tho explanation sight rather surprising. (In th case of a. sma,H rctarda.tion of current) is, that during t)u),t, iia-If of th motion in whieb th pt'ongs tu-c th most scparatcd, th acts in aid of thc proper eicctro-magnet recovering powcr duc to and so natnrally mises Hie pitc)). Wha.tc'vcr tlie relation rigidity, of phases t\vo parts mcnt ing may be, Hic force of thu rc.spectivc)y proportional To ti)c nrst and magnct to tho into n):Ly be dividett and (tisn)acevclucity t]ie sostainof pitch.

(or acculcration). powcr of th force, G5. TI)e

is dnc exclusi-vety to thc second the atteratioti

of rsonance, gnerai it cannot pbcnomenon thnugh bc exhaustively considcrcd undcr tbc hcad of onc dcgrco of freeis in th main referab!e don), to the same goncral prineipic.s. AVhen a forced vibration is cxcitcd in onc part of a. system, all the other parts are aiso Innucnccd, a vibration of thc same pcriod whose amplitudo on thc cxcitcd, bcing constitution dpends ofthe eonsidercd as a whote. But it not systum unfrcquently happons tliat intcrcst centres ou th vibration of an outiying whose part with thc rest of th System conncctio)i is but Joosc. In such a case the part in question, a certain limit of amplitude provided bc not exccedcd, is very inuch in thc position of a. systcm possessinf onc of frccdoni and acted on dcgrce by a force, \vhich may bo as ~e~, of thc natural regarded T)ic indepcndcntty pcriod. vibration is accordingly governed by th ]a\vs we bave ah'cady In th case of approximatc of pcriods investigated. to cfpudiry which t)ie name of rsonance is gencra))y th amplirestnctcd, tude cvcn In other considcrahic, may be very cases it titough bp so sma]! as to lie of !itt)c account; might and thc prcision

C5.]
in th adjustment required un tlic th effect. dpends Is subjcctcd. Among tuning, ciated \Vhcn bodies may
with

RESONANCE.
of th degrcc resound strctched as
Is sounded

611
in order to bring out to winch th systcm prcision of

pcriods

of damping without in an

bodics be

winch mentioned

extrme and

sounding-boards,

membranes, tho pianoforte in thcir

assostrings and thc violin. thcsc Thc

th
arc

propcr
caused

note

neighbourhood,

to vibrato

in

a very

note may bo made by singing nrst raised tlie con'csponding by any of ils ttrings, Iiaving giveu to any note bc plu.ckcd Or if onu of tbo Mtrings beionging dampcr. a Itarp with tlie nnger, its feHows will be set Into ()ikc string) cxperimcnt as may immediatcly vibration, T)tC piienotncnon of ~'hero n. vo'y accm'ato cases order to cHeit on rsonance is perfect across thc is sumcicnt and in tuncd wards
bc rsonance c([uality

nianner. perceptible tho into a pianoforte

proved
is,

thc nrst. by stopping mo.st striking howover, of pcriods is nccessary

tuning forks,

in in

t))c full cfrcct.

Of

tins

class

'muuntcd

boxes, are a. conspicucus thc vibration of ono fork

Witen thc UMison example. wIH be taJ~cn up by anothcr of a room, but thc slightcst of pitch width dviation to l'cnder thc phenomcnon almost insensible. Forks arc commonly pcr second that a dviation from unison ail thc dincrencc. used for thc oniy forks giving Whcn thc purposc, one bcat arc '\vcU backthcm must bc

of 25C vibrations it is found a. second and and ciose

makcs

a!ternatc!y, IMustrutions vitinn

th vibration togcthcr, may be transferred furwards bctwcen thcm scvcral times, by damping with a toucit of th nngcr. of tho powerfui c~ccts of isochronism

t))e exprience of every onc. Tticy are often of importance neld.s from any with which acoustics isconccrned. in very dinerent few things are more dangerous to a ship than to lie For cxample, in th trough is ncarly that (iG. Th of th sea. undcr thc innucncc
ofwavcs whose

pcriod viz. (1)

of its own natural solution of thc

ro)Hng. for frcc =0 vibration,

quation

M+ /C!t + )'M

form by cxprcssing tlie arbitrary conmay be put into another of intgration J- and a in ternis of tlie initial stants values of !< and Wc obtain at once and M, which we may dnote by

ONE

DEOREE

OF FREEDOM.

[f!G.

Th rc8u!t

cfTectof a.L tune

~7i.s to gcncr~to < will thcrcforc be

in time

f~' a vdocit.y

r<

w))o.sc

U bcing The gencrally On t = 0, and

the lowcr

force at time limit of tlie

< is so far arbitr.-uy, intgrais to make it zero. u aud as is givcn by but it will

bc eonvunicnt this the supposition complete

(G) vanish,

wlien

solution

Whcn Y~Ish

t Is sumdently on account oftl)c

grcat, factor

th e-~

tcrms tcnf] complementary and mny ti~en hc omittc<1.

<o

C7.] G7.
tho

TERMS For most

0F

THE

SECOND

ORDER.

G3

vibrations

:m m~nituly This ttons. oi' fur isoclu-ontsm forecd but

to consider it is sufRcicut purposcs uf t)i0 systons, with which wc may ha.vc to deal, srn:Jl vibrato Infiuitcly or i-iLthcr as simil.n' sm!).H, lim's of thc important rust.nctiu]i is thc i'omuhttion acoust.ical fur
t'r<jG vibrations, and

(jf

pc-rsistcucc

of

vibrations.

ordina.tc

on the s<trc t]ic rcmaindur dcvutc motion

so smaU

of nowevcr, ph<jnunicn:t, winch not insigniricant csscutiidiy charactcr, d<jpc!td Wc will thcrcforc and highcr ptjwc'rs of thc motion.

Thcru

arc,

pcriod a sub-

of

a that

syton th

to thu of this chuptcr of onc dcgrcc of freuttom, thc atid Ilighor powcrs souarcs for the

discussion

of

thc bemg

motion
can be

not

a~togcthcr

ueglefted.
The approximate expressions

potcntlal

and

kinctic

nergies

wIM be of thc

form

G4

ONE

DEGREE

OF

FREEDOM.

[G7.

tha.t shewing by its octave !'fti~Htud(.' octave 111

tho

propcr (2y:), whosc uf Yibr:d.io!t.

suu))()

thc

tone

(?;.) of thc

?'e~e A t'tihc(!

is accotnpauicd system lucrotses with tho importance

of a. tunitig )H(;:t.ns of a bow, imd wit)i thc of the octave ):(,tu]', t))c cxist,unco manifcst to any may bc niadc onc. thc same By foUowhtg inethod the ca)t appmxnnatioa hu ca)-)-iud furthcr but wc pa,.ss on now to the case of :). syston )n which thc recovo-mg with to power is synmiotncal respoct )hu position of cqmtibrmni. T])c quation of motiou is t!tcn app)'oxi)))n.tu!y to i-ufc-r to Mio vibrations uf a hcavy may be uuderstood or oi'a )u:).<1c:u-n<jd at tho end ofa, p'n(h)!un~ sti'iu~ht spring. If wu t:tkc an a jh'st M=-~ cos?~, cotespondi)]g npj~'oxitnatmn to /9 = 0, a.ud substitatc in tttc tcnn muhiplicd hy /3, we get winch

(.~r c~.ii guncr.diy th pcrccuc fork causcd to vibrn.te strongty by :ud cf i).pp!i:).uccs, to bo cxpl!uuc(t

Corrcsponding obtain ni tho grcatcr

Lun:n-

without Thcury,

last to the tcrmofthiscqnation, soh~tion a tcrm oftiie form <sin~, Jnnit with t. Tt.is, as in a paraUd indicatcs that our assumcd iirst

wc

shonid

becoming case in t]ic

approximation

is not rc!).]!y an at a]), or at Icast docs not coH~e approximation to bc such. If, Ilowcvcr, wo bikc as our starting point u =~4 cosM~, ~ith a, suitaUc vaillo for M?, wc sita find that titc solution tnay be cotnplutc() with thc aid of perio(]ic tcnns lu fact it is on!y. evident buforchand that all wc are entiticd to assume is that thc motion is approxinuttely with a pcriod simple harmonie, ahthe as if ~M'o.-n~n<< sanic, A very /3=0. cxamination slight
is sn~cicnt to s)tcw that the terni varying as M", not on!y

but tlmt

may,

~~M< a

afcct

tho

in solution, is assumed pcriod n)!)ttcr by tiow little, must at Icngth ccasc to rcprcsent with any approach to accuracy. Wc takc thun for the approximate
cqnation

period. which thc

At

tlie

saine

time

it

is vident

no wrongly, thc motion

ofwilichthc

solution

wilibe

67.]

TERMS

0F

THE

SECOND

ORDER.

655

in /3 thus produces two cS'ccts. It altcrs thc pitch of th fundamcntal a,ud it introduecs th <MeM!~ as vibration, a uccessary Thc altration of pitch is in most accomp~nimcnt. ca~cs excccdiugly on th square of the amplitude, smalldcpcuding but a It on it little, may is uot altogether insensible. little, that in thc pitch samo though very be remarkcd Tuning as th risc gencrally vibration dics away. of pitch dcpendence is of thc the approximation Thc result in tbat forks

Th

tcrm

occurs wlien tlie amplitude form M'M+mM, as may be seen by to th solution of (1) onc step furthcr case is

slight force of restitution continuing than (3).

Thc but in

difference one

w"

there worth respect noting, namely, in (8) m" is always that than while in (7) it dcpcuda greatcr on the sign of /3 whethcr its effect is to raiso or lowcr the pitch. In most cases of the unsymmctricat class the change However, of pitch would on a term of tho form M' and depend partly on another of the form /3 and thcn partly

is of the same is a distinction

order

in J. in bot)i

cases

C8. forced

We on an

now

pass

to

the

considration by two

of

the

vibrations forces

unsymmetrical

system

harmonie

Thc

cq~a.tion

of motion

is

CG Substit.uting

ONE this

DEGREE
term

0F
muJtiplicd

FREEDOM. by wc~ct

[68.

in tLc

vibrations rcprcscnt having frcqucncic~ arc scvcndty thc d.u.bius ~.<I tl.c sum nud di~n-nec of t)iose of thc prin~ncs. Of th two latter tlie ~e iUT.phtudcs to thc product proportion~ of the origine ~nplitudcs, s)icwing th~t t!iu derivcd toncs incrcasc ni relative which
impcrt:tuco with tho intensity of their p:irunt toucs.

Thc

addition~

tcrms

lu
cousequeuecs

future

wc shidt have to chitptL-r which Hulmlioltz lias dcduced

consider from this

thc

important

thcory.

CMAPTER

IV.

V1HRA.T1NU

SYSTEMS

IN

Gi;NEl{.AL.

G!).
of f), systcm

WH wu
Systems

ha.ve
posscssed

now
of

cxamitied
onc dcgrce

m some

dctfn!

the

osciH:t,tions

at

whicit

have o!'(!cr

an'ivud,
cnjoy iu

hve
guticnd

m:Ltt.;ri:d

of frccdom, nnd thc i'esu)ts, a vcry widc apphca.tion. But more than ouc dL'grcc of

at. any moment cou(1gur:ttio)i scvcnd vin'mbic must bc spccificd, uxhipcmtott qn~tttidcs whici), by :t ~(.'))ut':dix:t,tif)t) of ):u)gU!~c ori~hin.Hy cm]))oyc<l for a ponit, arc caUutt thu co-or~t'~f~es et' thc systcm, thc uumbcr of indcpcudcnt co-ordin:tt(js bumg tho MK~ q/rce<?o?~. Strictiy spc:dting, thc disphtccmuuts to {t n:).tm'a,l arc possibtc systcm infmitcly hu l'cp~'cs~ltc(~ as m:)d up of a finitc numbcr Viu'ious, and caunot of dplacements of sp<jcifiu(1. type. To thc cicmcntary pru-ts of a. so)Id body !uiy nrhiti~ry dispt~ccmcnts may bc givcn, subjcct to coti()Itioi)s of cotitituuty. It is ody by a pt'ocL'ss of idjstraction of t]tu kind so constiUttty in N~tuml th:it pr.LctIsctt ThMosophy, so!i<)s aru trc:t.tcd as )'i~'i.d, fluids n~ n.nd othcr snn. incompressible, mtroduecd so tli:).t th position of a, System cornes to phdctitions on :), finite numbur of co-ordin:).teH. depoid our intentiou to cxcludc thc considration infirntely
mtcresting

frcudoui.

In

to (tufinc

thcir

It of

is not,

however,

various

freedom

oti

that as Wc

of t!ic appHcatius direction. But such Systems of othcrs, wl)osc commence accordi))g)y by a, finite uumber

contnu'y, of this results

arc most

thc

Systems possessin<'f somc of thc most eh:mtc]' will lie in

limits sh:).ll

frcednm of

conccivcd conveuicutiy is of a, more rcstncted Mnd. systcm, independunt position co-oi'diitatcs -~r, wtiose

with

is spccincd &c. t~ 70. gation of

Thc th

ma,In but of it thc

prohicm

of

Acoustics

consists

itt

t!io

investi-

vibrations will

of

cquihbrium, st.itica.1 part

a, system about bc eonvenient to By thc

a position commence

of with K_o

stable the Vc-

subjcct.

Frinciple

of Virtual

C8

VIBRATING

SYSTEMS

IN

GENERAL.

[70.

locities,

if

we of

rcckol

th

configuration cuufigumLtuu

of t))C qud.cr~i.ic function t!~t the displacemcnt co-ordiuatcs, be sufHcicnDy provided smdt. Tins quantity is ciUlcd and reprcseuts thc work thf~t may bc in passing from the actuel to tlie gfdncd equilibrium configuration. We mny write

equilibrium, will bu :(. h~~m'nc-ous

eo-oi'dmatcs tlie potentiel

&c. energy

from

tho othcr

of any

Since

by supposition

quantitics c,c~, c, all real values of th eo-ordiaa.tes. 71. If tlie

equilibrium &c. must bc such

thc

is thoruughiy stable, that V is positive

tho for

bc Jisplaccd from tho zero configuration system thc action of given by new configuration forces, thc ho may found from thc Prineipic of Virtual Velocities. If th work done forces on th hypothetical hy thc given dispkcement 8~, S~, &c. be this &c.

must expression nro ludcpcudcnt, by

mincd

bc cqu:vn,!cnt to 8F, so thfttsmcc the new position of cquilibrium

8~, 8~, is doter-

-where

is no distinction in value bctwecn c,, and c, From thse quations the co-ordinatcs may bc dctermmcd terms of the forces. If ~7 bc thc dctrmIuMt

there

in

71.] Thcsc

RECIPROCAL

RELATION. ~r,, &c. uniquely, t. co.~idi.'t'Htiuit ~h'

G9 doca slaco Lh' 'f~f;{'.L'uh values of tha winch stable ia

dtermine quations ] .'i~ ~u t).ppcm. ilot Y:in)Hh, Mt ~ont 'iLb~til t! === 0. &c. could othcrwiso ttmt timt be

s~tisficd tho t!ie

by fmitc

co-ordina.t.cs, oniy provided to th hypothcsis contnu'y iu t)ie xct'o conHgura.tioa.

?'(t~'os wcre systcni

suitable,

is tboroughiy

If thc forces them alone equal,

~F, a.nd aud wu then

:i displacetneut

be of recoguiso of

the that

same

Mnd, we may suppose a force of any type acting

t.o tlie type cqual of tlie first type duc to thc action of au cqnid force displiicement of thc second For example, if and R 'be two type. points in ~ny maunc! the vertical of n, rod snpported dehorlzont:dty ricction at jl, whcn a wcight }F is ~ttachcd at is tl)u s:t.me as at 7?, wlien ~F is appiied at ~t\ the dtection produces 73. ordinates, Since F is a homogeueous qua.dra.tic function of thc co-

f), second

If wbich
thc

+ ~~).
neeessary

~+

~~)
forcus

~c.
n.ro

rcprcscnt ~+/

auniticr

(Iisp!act'mcnt

for

~+A~,&c.,thecorDeo., 1874, nud MMeh, 1875.

Ou thiB eubjoct., sco 7~tt~. J/<

70

VIBRATING

SYSTEMS

IN

GENERAL.

f73.

ruspouding

potentud

cncrgy

is givcu

by

whL'rcA'FistiKi'n~n.'ncuojf'thupntcntI.dcnL'rgicsinttxjtwo uoticu ciLScs, :unt wu must p:u'<.Icnl:n')y rui~iuu, 72 (I),

tIi:Lt

by

tlie

i-eeinroc:~

From

rclating

(~) and (~) wu to t!ic vainc of and first tu given thcorem

may fur

deduectwo

nnport:u)t

Dicorc'tns

a systeni respect! t!tc

subjeetcLt

to <Ivcn

dis-

pitt.cement.s, 7~. mcnts Th

forces is to

ycty. if
given ()isn]acc-

thc C())tti~u)':)tK)ii) (llot su~iclunt hy ttionscivos in a. systcm b(.' produd by f'urccs uf con'c'spundixg typ~s, t)tc rcso vaJuc of ~for thc :uid m u~uHi))rium, Hulting .sy.stcm displaccd, is ns sin:dt as it can bc u))(icr th givcn di.spinccmoit couditiun.s' and that the vainc fur :Uty othcr of excuc-ds tins couhgurattou by thc potcntia! uncrgy of t)m two. Thc on)y in undurshuidit~g what for cxampic, Suppose, which of thu cunHguratioR wincli is th In thu abovc statcmcnt diHioLdty is ntcant that thc by 'fores systum (tiSurcnce consists

eNfuct t!):tt, to dtcrmhio

agivcn point jf-* is to bu subjcct thu force of corrc.sponding !n(U)t; type is ut thc ])oint .P itself. And gun(.'r:dty, thc

of coi'r<spo]]di!)"' types.' is a. strutchcd of stri))" to an cbligatory dispJaccItre a. force

applicd forces, th by which is to bc tunde, must bc such as woul(i do displacumt.-nt proposcd no work on Hic systum, proyidud on!y tiuLt thut disptuccmcut wurctio~made. By a suitabic choicc of co-ordinatcs, ttic givcn vaincs
displaccmcnt

cotditicustnaybe ?' co-ordinatcs wdl thcn

cxpt-L-ssud

by ascribmggiven nud thu conditions

to thc

first

bc rcpr<j.s(jntcd thc foroja by inaking &c. vanish. rcfur to any ot)~cr contypL's f -+A-~ of thc systum, and ~+A~ bc tho corrcsponding hgnratiou forces, we are to suppose that ~c-. as f:n' as A~ aH vanisli. AA~, TIiusfor tite first r suifixes fur th vauishcs.aud remaimD~ 0

fm to thc forces of thc rcmaini))~

74.] sufHxcs
.P'r ~Fvanishcs. i.~ aise xcro.

STATICAL

TIIEOREMS. is zero, n.nd

71 therefore

AccordinglyST.AHon~c

2A~=~A~.A~(1), which othur that if thc provcs than tinj prc'scribcd givcn dplacements thu potcntial bo niadc any la incrcased T'of in

by t)ic encrgy By means

l'cl~xation m

cncrgy way, of thc diffrence of the configurations. of t!i!s t))corcm we may trace thc cH'cct on
sttH'ucss ofa. System, suhjcct to given

any cqui-

t!)e

displacemcnt

conditions.
librium

For,
connguration

ifaftcr
Le

tlic

altration

m stitTness thcvidnc of

thc

original

considut'cd,

t))crcto seoi, whctt ditions. of thc of

is by supposition Icss t)i:m bcforc; therc will be n. still furthcr dinunution tlio

Vco)'ruspon()ing wc h~vc just :md,as in tbe value of F'

the niterud conto cqnilibrimu undtjr Hystcm passes Henco wc condudc titat a. diminution I)i as a functiou in the actual vatuo co-ordin:t.tcs also n diminution cntails
a systcnt is subjcct to given disp!:).cemcnts.

F' whcn

It spokcn ends tbm'c bar.

will of be ap-

bo undur.stood may vanish*. l. For

cxample,

tluit

in

pa.rticuhu.'

cases

thc

dinunution at both a force will ofthe

disph~cud latcndiy piicd, thc potentiel cnurgy by :).ny relaxation (however 75. Tlic second

if a point J' of a bar dampcd to a given small antount by of thc loc:d) relates dcfui'mation in tite stiihess

be diminished

to f), system displaccd ~tM~ forces, and asscrts that in this case tho value of V in eqnilibriuni in is gi'cater than it would be in any other \vhich connguration thu syst~'m coutd bc maintained at rest undur t))c givcn furecs, by the opration of mcre constraints. We will shew that tho )'c?/MM~
ofconstt'aints increascs

theorem

t!)c vainc may by

of that th conditions of con-

TIio co-ordinatcs straint arc cxprcs.scd

bc so choscn

~=0, Wc va)nc hve of thon

~=0,=0.(1). ~P~ (1) by <c. !~))d. arc givcn, tho Thu second

to provc that whcn ~P~, V is Ica.st whun t!tc conditions dcnutud tu Hunce S~A~=SA~P=0, as bufot'u ')' inchtsive

bcing configuration that fur snrRxcs up sunixcs A~F vanislics.

+ A~, &:c., wc seo and fur higinjr vanishcs,

Soo n. imper on Goncrfit Thcorcma rchting

Stuady Motiouij. 2'/7. Af~ Mure! 1H7S.

to Eqnilibrium

aud luititd

nt)d

~2

VIBRATING

SYSTEMS

IN

GENERAL.

[75.

and

therefore

shewing stramts

that is cqual

thc

incrcase

in

F duc

to thc

rcmoval din~rcnce

of the ofihc

contwo

to tlic potcntial

encrgy of tlie

configurations. 7G. a systcm impulses. potuutm disptaccd, with th mdependcnt tlie system If ~partie e We now pass to the of thc initial luvesti~tion motion of which starts from rcst undor thc of givcn operation The motion thus is Indepcndcitt ~equired of any .vhicl~ the system encrgy n~y .vhcu possess actu~y siueo by tho nature cf impulses we h.wc to do only mitml itself configuration Thc initial motion Is also of any forces of Huitc kind, whethcion imprcsscd from without, or of the nature of viscosity. Q, 7i' bc the component to thc axes, on impulses, parallel ~vhoso co-ordinates rcct.nguhr h.vc by are

DAlGmbei't'sPj-iucip!o

in virtuc by the particle aud impulses, to auy correspond disarbitrary of thc system which docs not violate placcmcnt thc councction of its It is parts. to transform required an (1) iuto cquatiou cxprcsscd thc by independerit gcncralizcd co-ordinn.tcs. aequircd For th first side,

whcrc of the

dnote

th vclocities

whcrc presscd

the

kinetic

as a function

cnergy of

of

the

system,

is supposcd

to be ex-

&c.

76.]
On tlie second side,

IMPULSES.

73

whcrc
dtermine

8~,

S~,
tlie

&c.
motion

arc
wc

now
ha.vo

eompletely

iudcpcudcut.

Hcuco

te

whcrc S ofi;i)putsc.

&c. may bc constJcrcd

as tlie

gGneraItzcd

componcnts

'77. Since y is a homogcncous ralized we may takc co-ordiuatcs,

quadratic

fuuction

of

the gene-

whcrc

there

is no distinction by the nature

in value

between

0,, and

Again,

of T,

to that of th analogous of a. system froni a configuration of stable cquitibrinm displaccmcut lu th prsent by steadily kinetic a,pp!Icd fores. tucm-y th initial T bears to the vclocities aud impulses tho same relations encrgy as in thc former F' bcars to th displacements and forces respccttheory

The

of initial

motion

is c!osc]y

74 ively. complet~ If' for

n. systuin

VIBRATING In respect in!).s)nuc)i. as a,

itomogcuc'jus

SYSTEMS the thcory

IN

GENERAL.
inot!ons is

[77. the more

one

of initi:d w)ti!c

is cxactty,
(~uadrntic

approximatcty,

dunotc st:u'tcd

nnc set

gt'ncmt oniy fuuction of th variables. of vclocitics and impulses

a. second

in

sct~ wu iua,y pt'uvc,

f'run) rcst, :utd as in 72, th fuDuwing

recipt'uca.l t'L'ta.tiun

This

thcdn'm

ndniits

It is kuu\v)), :utd will th:)t thu tnoti~n ui' starts i'rota

cfitttcresting' bc provcd I~t'jr

to f)~)i<~ motion. :)pplie:ttl<)!i in thu coin'sf.! of tins work, which ineotnprcs.sibtu liquit), t))!~t its cumpom'nt vutoeitics

!), inctiun!uss

rc'.st, i.s cf such :t. kind ~t nny point aru thc con'L'spondi)~ dit'fu)'c'nLi:d cocHicicnts uf n, ('(.')'t,:nu fnncti"n, c~Hm] thc vctocity-potcntiid. Let t))c fh)i() bc sut In )n(jt[on (h.'tL'nnincd bL-in~ by :t prcso'ibu)! /S' of :t c)')SL'() spucc describud hy thc normid s)t:n'L'd by t!ie Hnid ()L'fo)m:).tiu)i of th(j surface tn'bitt'in'y mution within it. Tiiu rcsniti))~ is of vctocitics of th cloucnts winch, in contact witti thcm, m'c duuotcd by \vltich Hunce
secoud

if M be tho vc'ioeity-potcuti.t.], notes tnc ijnputstvc uf

suifacc pressure. u,

luto'prctcd

phy.sica!)y If to

d-

by thc

t]iC(H'cm,

bc t]io unother

VL'Iucity-potuntlid set uf arbitrary

motio;),

corruspuuding

vclocitics

quation bcsi()<s~'thurc prusunt ruHty. ]m;t)n)d yur

an

immcdiatoly be ou)ytixcd unabius

from foiJowing soli<[s inunur.scd

Grccn'a in tho

if thcorcm, ftuid. Th Itighur gcnu-

us tu attributc

to it a much

untm'rscd (.x:unp)u, t)ic or ni part., m:i.y t)c irc-L', :dtogct))cr upo)j tl)u)n by thL! Huid prcssm'

of buing soHd.s, mstuad Hxc< to takc tho motion iinposcd

78. spcial

cnsc A paTtk'ular notice. In thu nrst

nf t)ic motion

gcticra! Jet

thcorem

is

wortl)y

of

78J In
ing type,

THOMSON'S words, if, by the

ordcr mcans of

TIIEOREM. impulse
onc co-ordinatc

75 of the
bc

a suitabic
of

correspondimprcsscd

n givcn

arbitrary

vclocity

on

a system,
in

imputse
to prcvent

corresponding
it from

to
changing,

M second
is t)ie

co-ordinatc
samc M

nccessary

would
\vo'c

bc rc'ptircd for the first co-ordiuatc, If titc given on ttic second. hnprMS.sud As :t simple ted the eMC of two sphres uxampic,
in a liqnid, wliusc ccntrca arc f)'L'c tu !)iovL;

velocity and ccrtiun will the as if J~

nmncrHcd

along

lines.
natnndiy

Jf

~t

bc

sut to to

in movc

motion also,

with Thc

i]))pt))su thc functions thcrc

bL'gin rctmin.'d

givcu thcorcm uf :uul

-B vulf~'ity, :LS.surts th:Lt if) thc this s:mic cvcn

of yt

prcvent thc motion !md 7? wo'c cxchimgcd

rigid bodius, C', D, &c., in the ituid, citl~cr frcc ill whulc or i)t part. Thc case of cicctric cnrrcnts cach mutually i)iflnencing induction is simihu-. Lct thcru bc two circuits prccisciy m titc ncig!ibour)tood of which thcrc may be a.ny numbcr wirc circuits or sohd condnctors.
tho

Le ot])cr

thuug)i or fixcd, othcr and of othcr snddcidy ill been by

If a unit
clectromotive

cnrrent
Impulse

bc

in thc circuit J, duvulopecl is the slulc as there would furcibly 79.

are

induced

have

bccii

iu ~1, hn.d tlic currcnt

dcvclopcd Thc motion

by

in of a system,
mcans of thc

on which
ncecssfu'y

given
Itnp)dscs

a.rbitrary
of

vclocitios
t)ic corrc-

nnprcs.scd

sponding Tiiom.son.

types, Thc

posscsscs conditions
vanish.

rcmarkabtc arc that

prnpcrty
&c.

discovcrcd arc
currcspoiid

by givcn,
to

Lct

thc

actu.d

motion;

and

~+A~, to cach of thc anothcr snmx

motion

~t.A~
satistying

~+A~, thc
vailislics.

~+A~ For cnergy

saine

cithcr

supposcd

AT~- u)' motion,

conditions. velocity New for t)iu kiuctic

2(~+Ay)=~+A~)(~+A~)+. =2~'+~A~+~+.

+ A~
But by th

+ A~.
rcciprocal ~A~.+.

+.
=A~+.

+ A~A~+ A~A~+.
(4) of'77

rctatioa

of \vbich

tlic former

by ItypoUtcsis

is zro;

so that (1), J

2A2'=A~A~+A~A~,+.

7G

VIBRATFNG SYSTEMS IN GENERAL.

[79.

that the shewing th actual motion

c motion bc c.nrlod actualtymduced

of the snpposcd encrgy motion excceds that of f th of that motion by energy winch would hve
..ith in thc },t. System rc p,.d.~ bas tf~.s ihc Jcss fbnner. oucrgytlm,i Th

s~. yn.g tho same velocity conditions. In a snbs~.cnt we shall ch. ptcr make use of this to find a propcrty Jinut supenor to the of a set in motion cncrgy system with prcscribcd vc-Iocitics be made in th ~ny dnnmutiou inertie of of t)je parts of any a system, t)ic motion to prescribcd corresponding conditions velocity wu iu genem a Thc value of undorgo change. will nece.ss.riiy be less than for t)~ere wouM be a before decrease cven if tlie motion rc.nained and tl.crcforc unchangc<I, w]~en th /b7~ niot.on ~s such as to make 7' an absoJute mim.num. Converselv m tlie inertia any incre~c increascs thc initia! value of T. lu. thcorcm Is to that of 74. analogous Th analogue for initial mot.ons oi th thcorem of 75, relating to t].c potential of ~~gy a.system d~.ced by given forces, is that of Bertrand and may be thus stated -If start from rest under ,y, th of opera.on givcn nnpu!scs, the kinetic encrgy of tl.e actual motion Lxcceds that of any otlier motion which th system I~.ve might been gu.)ed to takc with the a.ssistance ofmere constrain~ by the kinetic encrgy of the din-crence of t)to motions' 1. 80. WcwiIluotdwcUatanygreaterIengthonthemcd.anics to impulses, system subjcct but on to pass investie for continuous equations nation. Wc .shalt suppose Langes the connections that th bniding togcthcr parts of U.c..svstc.n are not functions o.plicit of t). sucli tune; ca~sof H motion as we shall have te consider will bu .shcwn to speciaily ue wiHun th of the scope investigation. of a combination Ydot~ V c10ci tics, (~~ with that of Virtual

~y

other

+ y8~ + ~~) = S (.Y~ + F~ + ~) dnote a d:sp!acemont ofthe of th most system ~herc 8~ t~ -nection. of":f ~r Sn.cc r'~ parts. th d,sp)acemcnt.s of th individu.-d of partides t arc ~nutuaHy system are not indcpen~t. relatcd, T) .ow is to transfonn ohjec tu other variahJc.s which s!tatl bc indcpcndent. We hve

ThomBou

auj

'fuit.

~il.

~,7..V.y.

Mareh,

1875.

80.]

LAGRANGE'S

EQUATIONS.

77

so tha,t

if

as n. quadmtic cxprcsscd ni gcuerfil functions cocfHcicntsorc

T ho

function of

of AIso

whose

Since

~F8~

denotes

Hie ma.y bo

work

donc

on as

tho thc it

system gcncralized is

during

disp~cemcnt of force. ponent In separate tion of may th tho case of from system. thosc

recoarded

comto

a, conscrvativc

system

convnient

parts which dpend only on thc connTurfiif V dnote thc potential Thus, wc encrgy,

write

whcre are not

~P is now aIrGady

limited taken

to tlie account

forces of in thc

acting tcrm

on thc ~F' a~

system

which

78 81. it is Tilcrc ofton

'n.i

VIBRATING is also

SYSTEMS another to
yip.Hy. is rctiu'dcd by

IN of

GENERAL. forces spccifdty, whoso

[81. existence thosc

c~h

group rccngnizc
if

a~/fmt~gcous
fn.-n~t thc syston

namoty
L),)H

?;r..s)r~

y~ forces

.pp~o

piu'ticlc eomponbnt
quation

of

t)ie cH'ect will bc velocitics, to (1) 80 by t!tc addition

to proportion~ sitown in thc fu)jd:uncnt:U Jcft-Jtiuu! mcnibur

its of

tl)c

thc

terms

whcrc

A-y,
pos.stbiy

but

tr:msibr)n:itn)u

cHucted

iu

t))o ve!(jcit.Ics, on t)tc configuration of thc syston. dcpcudcitt T))c to thc indc'pcndunt co-ordinutus &c. is a. sirnihu' manner to tJKtt of indpOKicnt

nrc

cocfHcicnts

of

is hl.c 7~ it will bu obscn'ed, a honngcncons quadratic fucetiou of t!tc vuioctLies, for :dl rL-:d v.ducs po.siUvu of tho v:u-)ab)cH. It !~pruscnt.s hait thu r~tu ~t whidt cncrgy i.s (hs.sij~~cd. Thc abovc i-ufcrti to t~tarding itivcsti~tiua iornus propordonat to thc absolute but it is equaUy vclucitics tu cuusidur important sucb as dupend ou tho p-e~~c vulocitics of thc parts ci' titc and furtuuately system, tins eau bc done witiiout auy incrcaso 01 complication. For cxampic, if a furcc aet ou the partielc xi to thc-rc proportiona! will bo at thc samo momont an force acting cqu~I and opposite T!ie additioual ou th partide aterins in the faudamental wi)l hc of the furm eqoatioti

and

so

on

pfirticics. whicli still

of mutually pairs TIic only effect is th addition of ncw iu the form (2)'. We silall secappears any

for

numbcr

of

]nfhtctic;ng tcrms to 7~ t)iat prcscntly

Tho difforecoes rferred to iu tho toxt may of course pass iuto djilcrcntia! eoefUcients in tho case of a body oontiuuouBly deformed.

81.]
thc existence of

THE

DISSIPATION

FUNCTION.

7!)

7~ which may bc cailed th DiscertiLin rctations Funetion, sipation implics among thc coenicicnts ut' tho gcncralizcd of vibration, which cqn:t.tio!]s ctu'ry with Utem l, Iniportaut couscqucnecs'.
But althougli In an important c]~ss of th c~ses

tho

fonction

thc

cffccts
romains

of

viscosity arc opcn whctitcr cases. cannot minc sono tbat

l'ept'cscuted by thc function snch a method of rcprcsGntation 1 think it pTobable th~t it is so; but to provc any cxpect gncmt proprty of n strict (L'nnition
are symmetry

question

is apptic:).b)c in aU it is cvidcnt that wc of viscous forces cnable sun~'Icnt as dunvcd tbc of rctarding vcloeittcs
of

!) t'hc absence

\v!ncb
viscous

will arc

us to dutern.rc not. to from forces of thc tlic form In shcw a

wit)). certainty wha.t forcus of CMCS cons!dc;['!Ltio)is th forces At

to

!H)d what

retardmg fnnction. disHipatioti arc proportional parts uf tlic systuia,

ma.y bu any rate absolute liavc

rcprcsoitcd whuucvcr or relative

thc

wc slutti

quations

motiun

tho condition that t))0 motion mny now mtroduce takcs of a. conH~u'tt.tIon place iu tho nn)nc()i:).tc neigh'b(n)i'hoo(L of t)iorou~I)ly stable cquHibnum 7' and F' arc then homogcncous functions of ti~c vclocitics witli coufHciunts winch qmuh'atic aro to bc tatcd as constant, !ui(l i.s a snnUar fuucttou of th co-ordina.tcs that to bo tticnisdves, provided (as we suppose t!io case) with the fuuctious :n-c ofthc heconic the origin of CMh of co-ordmatc
cquilibrium.

82.

Wc

couhgura.t.ion arc ossentiaUy ordcr assumiug

positive. ofstnMil

Since

second h)iear,

quantities, the form

to con'esponj Moreovcr all threo ~V ternis of tho form f/ the equations of motion

is taken

whcrc not

under alady Tho

~P arc provided

to

bc for

mciudcd by tlie

ail diffcrcutial for tho

forces

n.cting coefficients

on

thc of

System Faud

a pnpor o/' the ~~tCMNttca!

Funetion Dissipation on Gonoral TItooroma Soete~

ttppoMs retatmg

Rrat

timo,

for Juno,

to Vibratious, 1873.

so far as 1 am nwnrc, iu m tlio 2~ocfe~HM< publishod

80

VIBRATING

SYSTEMS

IN

GENERAL.

[82.

Thc

threo

quadra.tic

functions

will be expressed

as foUows

1 whcrc thc cocfHeicnts quation c are constants. (1) wc may of course fait back ~and F; or .Fand T, to vanisii. of intcrcst in tlie on prcvious

From results

by supposing

A thin!

set of thcorcms

tnc.tymayboobtaiucd but it is uuneccssai-y If we substitute we obtain a system

byomittlng~and to pursue the subject thc values of T, F and which may

to E)~appHcation F; wliile ~isrctaincd, hcrc. F; and write bc put into D for dt tlie forni 1

of equations

we may further, draw an important inference from of our equations. If correspondinc to tho two sots of forces respectivoly H\ 0 i l't~tt"' \P motions dcnoted by be possible, thon must also be possible thc motion in conjunet:on ~,+~ ~+~ with hc forces ~+~ a p.rticuL ~+~ Or, case, when there arc no impressed forces, th superposition of any two natural vibrations constitutes also a natural vibration This is thc ccJcbrated of thc principle Coexistence of SmaU first Motions, cnunciatcd clcar)y by Daniel Bernoutli. It will be uuderstoo.! that its truth in gnerai dpends on tlie justice of th a.s.sumption that the motion is so small that its square may be neglectcd proccoding th

83.

Before

84.]
84-.
l~qu:J toI' lui'~1~

COEXISTENCE

0F

8MALL

MOTIONS.

81

To invcstig~tc
.n:d we

th
4Yjjl

free

vibrations,
~jt.l!

wo must
Il f.:ytt~.m

put
On which n0

P-01H11IPl!e

thc coefRcienta for which therefore equ;)J t.~ forces -n!~ frictioual ~ct, v-'c wiH cotmttt'ttce~'itLany'ttf'mouwhic]). We havo M)~ functioDS of th symbol

&c. are

From

those

quations,

of which

thcrc

arc

lot all but of liberty, system possesses degt'ces Thc result, wliieh is of the samc bc climin~tcd. the co-ordinate ret<uucd, may bc writton ~=0.(2), where \7 denotca th determinant

as many (??t) as th onc of thc variaMes form whichcvcr bc

and Let
quation

is (if there i\
in

bc uo friction) \t Then by the is

an even

\ D.

of D of degrec 2M. coueidered as au roots of V=0 of dliferential th theory equations

function

most

genera.1 va,!uc of

~4, J/, J?, J~, &c. are a.rbitrn.ry constants. quantities but tlie consta-nts This fonn hoids good for eMh of the co-ordinatcs, In fMt if a arc not in the diffrent indcpendcnt. expressions whcrc the 2w particular solution bo ~=~ the ?'a~M ~t~ -~a. ~=~' M'c complete]y &c., determined by th

quations

where

in each

of the coefficients

such

as

is substituted by the =-~3' condition

for D. that to for of

Equations is a. root th th root othcr

(5) arc necessarily cc~upa,tible, of \7=0. Th ratios ~1/ arc tho pairs samc as of roots X~,

correspouding but the ratios ~1~ ~1, are distinct &c. titcrc Systems G

ratios. R.

82

VIBRATING

SYSTEMS

IN

GENERAL.

fgg.

with which wo arc doalinrr system an importt imposes restriction on the possible values of If wcro .6a}, elthciwoutd bc re~I and positive, or and wo sho-Jd obtain a particular solution for which tho co-ordinatos, aud with them th kinetic denoted energy bv

85.

Tho

nature

of

th

a. motion is obviousiy fdr Impossible wbose whoJc energy system, can uever di~cr from the sum of tho and kinotic poteutial with which it was energies Mimatcd at starting. This conclusion is not cv~cd by takmg beeausc we arc as much at ngative, to trace th motiou liberty bMkwards as forwards. It is as certain that t!te motion ncvcr ~s as tliat it nover will infinite, The same argument excludcs t).c of a possibility value ofX. complex Wc infer that aU ttie vaincs of are purely corimacrniary to rcspondmg values of ~a~e t)ie tact that Ana)yt:caHy, thc roots = of 0, considered as an quation iu are at! real and must bc a negative, of thc relations consquence bctwecn subsisting th coefficients virtuo of fact for all real values of the variables 2' and F arc positive. Thc ca~e of two degrees of liberty will be afterwards worked out in full. 86. Tho form of tlic for TIius solution may now be ~IvMta~cousIy ~d ~dng

incrc~c without a conscrvative

limit.

Such

changcd by wnting new arbitrary constants.

&c. (wherc

.=~1),

where 111n' found th ratios This

&c. are by writing ~1, ~1,

th

roots

of

th

-M" for in are dctcrminatc solution of the

of equation decrec = 0. For each value of and real.

complte tions of a conscrvative may be resolved mto difforent each priode others. such motion, as to reducu itsdfto If tbe

is thc

of tho frce vibraWe system. th whole motion normal harmonie vibrations of (in gnral) of which is entirely of tbc indepeDdcnt on thc original depending bc disturbance, onc of thse ~.), wc hve problem sec that

]
t'

NORMAL

COORDINATES.

83

on the constitution of th AI dpend and only thc absoluto system, and phase arc arbitrary. amplitude Th several co-ordinatcs arc always in similar (or opposite) phases of vibration, aud the whole system is to be found m the configuration of equilibrium at th same moment. hre the mechanica.1 peroive foundation of tlie suprcmacy vibrations. If the motion pf harmonie be sufHcientIy small, tho diffcrential becomc Iluear with constant quations coefficients ~hi]e circular functions arc th ouly oncs which (and exponentia)) reta-in their type on diffcrentiation. 7~ pcriods of vibration, determined by t!ic quation 0, are quantities = Intriusic to th system, and must corne out t.he same whatever co-ordinatcs to define the conmay be choscn But there is one system n~uratton. of co-ordinatcs, which is that in which especially suitable, the normal namely of types vibration arc defiued of aU tlie co-ordinates by th vanisbing but onc. In the first type the original co-ordinatcs &c. Iiave let the quantity given ratios values be < so nxing thc absolute that in tliis type each co-ordinate is a known of < So multiple in thc second type each co-ordinate as a known may be regarded of a second and so on. multiple quantity deterBy a suitable mination of th quantities &c.. ~y configaration of tite as compoundcd system may bu rcpresentcd ofthc ~t configurations of these types, and thus tlie quantifies <~ thcmselvcs may b'c Jookcd as co-ordinates tite configuration upon of th system. denning Titcy are called tlie ttor~a~ co-ordinatcs. in terms of thc normal expressed ?' and V co-ordinates, arc reduced to sums of squares; for it is easily sccn that if the also appcarcd, the resulting products of vibration quations would not be satisned by putting of the co-ordiuates any ~-1 cqual to whilc thc zero, one was finite. rcmaining We from might ZD hve commenced with this that any two AJgebra homogcncous bo reduced by linear transformations transformation, 1
quadratic

where

th

ratios

We

87.

Th

When

assumin~ 0 functions can Ttms

to sums

of squares.

whcrc

thc

cocnicicuts are ncccssarily

(in

which positive,

th

double

sufHxe.s

arc

no G2

tono-cr

required)

84

VIBRATING

SYSTEMS

IN

GENERAL.

[87.

88. tlleorem Th
about

The

interprtation of considcrabio
cousorvittivu

of thc importance,
system cquiHbrium

quations which
vibrating is

of may
i)i

motion bc

leads thus

to a

statcd'. t. type when

period of a position

a,const)-!nned

of

st<).h)c

stationary

in v:Uuc

th

We might type is norm: provc this from the ori~inid cqua.tions of vibmtion, but it will bc more convcnicnt to the cmploy normal co-ordinatcs. Thc constnunt, w]nc)i may bo snpposcd to bc of such a cha.racter as to ic:ws only onc of fj-cedom, is dcgrc'e th quantittes represcuted in givcn by taking rutios.
If wc put

of th vibration gives thc period of tlie constrained type and it is vident tliat thc period is stationary, when a!l but one of th cocfncients that is to say, -when th ~l,, ~1, vanish, type coincides with one of those natural to the system, and no constraint is necdcd. of this tlicorem wc may provc that By means an iucrease in the mass of nny part of a vibrating system is attendcd by a proof all tho natural longation or at auy rate that no periods, pcriod can be diminished. tlie incrernent Suppose of mass to bc infinitesimal. Aftcr th altration, the types of free vibration will in be changed; gnral but, by a suitable th system may constraint,
r~~c~t')).?)! of ~;f; ~~<fma()ra! ,9of'~)/, Juno JH73.

This

88.]

PERIODS

OF

FREE

VIBRATIONS.

85

bo made to it is certain
whosu

Only the nodc Tlic straint,

If this be donc, types. any one of tlio fonner a motion which involves of th part thnt any vibration increased will I)ave its period inass lias been prolonged. whcn a load is placed at as a particula.1' case (as, for exampic, rctain
string)

oi' a vibrating tlicorem now allows and tlie

eau th

us to assert

romain period that tho removal

unchangod. of tlie con-

th of type, can only aScet consquent change in th and that therefore of thc second order; period by a quantity before the change. bc Icss than limit the free period cannot By ofinass must proloiig incrcasc wc infcr that a imite the intgration of th part Involvcs a motion vibration which of' every period but in and that in no case can tlie period bc diminishcd aliected, it may of th two sets of periods, order to sec the corrcspondcnce madu by stcps. be necessary to suppose the altcrations Couvcrsely, vibrating vibrations. In iike a system th efect must bo of a rcmoval to shorten of part of thc mass of all th the pcriods of a froc

manner

we may

prove

that

is that the potential change energy is unaltered, while th kinctic diminislied, energy of a given motion and convcrscly. arc aU increased, the periods of th free vibrations the effects 6f This proposition be used for tracing may sometimes of a constraint that th potential for if we suppose energy of constraint the condition gradually configuration violating tl]e in which a state of things we shall incrcases, approach of completeness. is observed with condition degree any desirod becomes each stop of th process every free vibration During any of th free pcriods more rapid, and a number (equal (in gnral) Th small. become to thc degrees of liberty infinitely lost) th powithout be rcached same practical result altcring may woftOM of any the kinetic tential energy energy by supposing In this case without limit. to incrca~e the condition violating finite th but one or more become large, infinitely periods at whcn tlie same as those arrivcd are ultimatcly th periods in one case the pcriods although potential energy is increased, have and iu tlie other been throughout diminishing. increasing, of making th altrations This shews the nocessity by example steps; of tlie under otherwise two th sets head wc of sliould not understand tl)C illustrations Furtlier pcriods. of two degrees of frecdom. eorrespondcnce will bc given

if the system undergo of a given configuration

sucli

86

VIBRATING me~s of wc ~h. th n..y

SYSTEMS

IN

GENERAL. value of

[88. tho frec duo periods to J~ any

By

principe ea.i!y

that calculato

tho

stationary <!cv~n a systc~ tit.es exaiaplcs tins work. 80.

y.steiu

eorrcctions

hypothct~ th dcpcndmg cf suci~ punod ou

type so

of found

yibr.tion wiH di~r of will

that from ti.e Le

prope~ the

to truth

th by

sl.nplu quanScvcral

tho

~uares

u.rcguh.-Itic.s. gtvcn in th

c..dcu!atiuns

course

of

Anothcr
vihr~ng

po.nt
,n .n

of unpor~ncc
~rbitrary type

reJ.ting
rcn~ins

to thc io

t
~c.sp

.ppcars
n.

~.tcst o
c ntuu~.s

of fhe p.riod to ~ny hypothcti~I u~ type is inciu.Icd and Ic~t of thosc n.tur.I tu t!~ system.
deior.n.t~n,

from

(2

88 that

of period be noticed o thc c~ bctwocn In thc

thc

vibration

thcrc cxcccd tLer.f..c that ti.

is

no ~i

I~t

uatu~

pericd;

h puthet

c.d"Y type typ.

c.uinoL Whe.

any to th

hy-

bclo~l,~ cLjecti.J~I~

Gr~est

~cdr' will como out t tao small, ~1 ohjcct

w~thout

of calculons result

usc~h~ t)~ uscd

he donc

ben,g
too

to
grc.t

approach
sacrinco

type jadgn~nt must bc th truth as nearly as can

of .hnpHcity.

hc.vily ~i,ht.d ~ht thc extrme case of an innnite ~hen tho Joad two of th .ould Le str~ht. st.~ As Je.~pl~ p~ cale..tion of tins Jun~ of which the rcsult is known, wo ~~h:d will w~th tcusion t~Tj~ anj 7 what mquirc the would period be on certam as to th supposions type of vibration. the origin of .r at t)io ~idd!o Taking of tho lot the string, curvc of vibration on thc positive sidc bo

ypcor Le tdu

~g froin

ncg~vc not !c.ss than bc~g whcn ihat y vanishes for 2' aud prcs~.s

~ul

on

thc

side

the

.nity. ~.=1 a.d it

of Im~c This form Wc will h~vo

tins satires now

in

Le su~icicut

of y th condid~ to form the exte c~t~

the

axis

89.] positive (tcuHtty, lialf

PERIODS of tho

0F

FBEE

VIBRATIONS. Thus, p being thc

87

string

only.

longitudinal

and

Hcucc If M==l, in its middie thc string vibratos point, and as if tho mass were concentratcd

Tho tho

truc

value

of p"

for

the

gravest

type

is

TT-T ,r, plu which

so that is too of p", gives

assumption small in thc as givcn by

form gives a pcriod of a para-boUc ratio 7r 1. Tlie minimum ~/10 or '993G (2), occurs when VG +1 ~==l'72-t74, and

It choicc

will of

he

seen

that

there

is

considrable

latitude

in

th

a type, even tho violent that th string supposition as two a period vibratos less than ton straight pices giving in error. And whatever tlie type wc choose to take, pcr cent. from it cannot be greater than the truth. period calculated and types of periods of a given system vibration is usually a matter of grt di&culty, from th fact that th functions to express tho arising necessary modes of vibration of most continuons bodies are not as yet rccogrigorous nised in analysis. It is therefore often ucccssa.ry methods of approximation, t!io proposed referring to fait system back on to somo 90. The determination of thc

88

VIBRATING

SYSTEMS

IN

GENERAL.

[90.

to analysis, and calculatiug on the supposition that the differerce between the tv.'o sy.ste~.s .aU. Th. r~ ~proxi.c!v is thus simple systems one of great more especially importance, as it is impossible in practice to realise tlio simple actually forn,s about winch wc eau most casily reason. Let us suppose then that tho vibrations of a simple System arc and thoroughly that known, it is required to investigate tho.sc of a systcm derived from it by small variations introducing in thc mechanical factions. If &c. bc the normal co-ordinates of tho original system,

of a character corrections depending

other

more

amende

and which

for

th

varicd

system,

rcferrcd

to

the

arc how only

same

co-ordinatcs

approximtttciy

normal,

in

which In eert.m cases

&, new coe~cnts equa,tious

small

quan ~cs. so t!.cir if ijagrangian

co-ordinates but may appe~ must bc small. From (1) ~c obtam for the of motion,

.In are erd~na

the

thosc

original systcm .h.ch corrc.spondto Let Thc in us say vanish. altcration variation

the

fondamental the variation our attention while a!I system tlie

e .1 a timc.

types of but on thc

of

vibration

fix

involving ordinates cntail an under ne of h' tlie

of

a single coone of them, co-

in tlie change tlie iund~c.tatcr contemplatcd e.pre~cd by the in to will

remnining ,vi!l in ~1 types;

normal alteratio~

cu.cumstanccs normal type is small.

but small.

other of the

synchronous variation but ratio of any known, ~e normal

mode

aftered

systcm

be dc-tei-mincd.

90.]

APPROXIMATELY

SIMPLE

SYSTEMS.

89

Since tha) with cn.ch diff'erential

th

wl)olc

motion

o~-ordinato

is simple harmonie, va-ucs a~ cos~, a"<! for D' In th

we may f.~Lft.it.utu

suppose thc occurs

quations tiio Snitc coefEcient

a"' quation

Th sincc small

otlier the

tcrms

a.rc to be Hcnce

neglected

in to ~)

a first and

both

co-ordma.te

(rcla.tivcty

approximation, its coefficient arc

quantities.

Now andthus tlie required If th result. kinetic energy alone undergo va.ria,tion,

The equation

correctcd of (2), not

value

of the

hitberto

is determined period used. We may write it,

by

tlie

?'t!)

without allowgives tlie value of p/ calculated ance for th change of type, and is sufficient, as wc have aiready wheu thc square of thc altration in the system proved, may he neglectcd. The terms included under thc symbol S, in Th which the summation due extends to th to ail values of of s other than type and are of th sign of any term ?, and a,, are positive, that of > p~ that is, if the mode depends upon p~. 2* If s be more acute than the mode r, the correction is ngative, and makes tlie calculated note graver than but if the beforc; mode s be thc graver, th correction ra-ises the note. If t' refcr change give thc correction second order. Since r, the

first

term

90

VIBRATING

SYSTEMS

IN

GENERAL.

[90.

mode of th system, gravest tho whole correction is and if r refer to th acutest negative; mode, tho whole coiection is positive, as we have aJrea.dy seen method. by another 91. take is not hc will tho th
quite

to

the

As case

an

of example of a stretched If x displaccmcnt, by

the

use

of these

formulae,

we

may p t

constant.

wliose string, ho measurcd t!ie

longitudinal from oue

transversc

he exprssed

configuration

density end, and at any time

the of thc string. being longth arc tlie normal co-ordiuatcs for p== constant, and t)iough hcre p is not strictly tlie configuration constant, of tbc systcni may still bo expressed of the same by means Since the quantits. potential cnergy of any configuration is tlie samc aa if/)= For 8~=0. constant, tlie kinetic we liave cncrgy

of tho velocities products would dis&c. arc, on that the normal supposition, co-ordm~tcs. As it is, tlie mtcgml uot actually cocaicicnts, thoug!i arc small evancscont, Lot p=p.+~; quantities, thcn in our notation previous constant,

If p wero since appear,

thc

Thus

thc

type

of vibration

M expressed

by

or, since

01.]
Let us

EXAMPLES.

91

to calculato tho disp~cemcnt of th apply this result mode (?'=2), which would of the second bc iu the nodu.1 point In the neighbourhood woro uniform. if tlie string of iniddte, value ofy is + &c, tho approximate this point, if x ==

Hcncc

when~=0,

approximately,

where

To the

show

the

Irrcgularlty at x = though JIrectIy. We

of these n.ppUca.tioji to consist in a. small thc result might

wc may suppose formula, load of mMS p~ situatcd bc obtained much more easUy

have

from 'l'lie

which rcal

the

value

of Sa; may bc calculated very simple.

value may

bmckcts

of 8x is, however, bc written

by approximation. Thc series within

The

value

of thc

definitc

intgral

is

and thus

Todliunter'a

f)t(.

C'tt~c.

255.

92

VIBRATING

SYSTEMS

IN

GENERAL.

[91.

th periods of vibraby equ:ttin~ tha.t of the loadcd parts string, part buing ca.Icutatcd on the of' :),pp!-oxim:ttc]y assumption unchanged type. As ~u cx:unp)G of tlie formu!:), ((!) 90 fur th pcriod, wo a, smaH lo~d may tn.ko tho case of a. striug at its c:u-rymg middie Wc havo point.. rcadily proved of th

as may also be tion of the two

and

if P, bc th value corrcsponding ?' is evcu, r is odd, and wheu = 7~ t)ms,

to

= 0, wc g'ct

whcu

whcrc

thc

summation If?'=],

is to

bo

extendecl

to

all

t!)e

odd

vducs

ofNot,herthan?'.

g!vlng square In

pitch of th ratio the

t~o

of .

the case

gravcst the value

tone

accuratcly correct

as as

far fur

as as

thc thc

gencml rtrstorJcriu~p.wiIIbc

of p,

throws grcat Hght on expansions thcory of vibrations of arbitrary functions in sries of other <\mctlo)is o(' spccif]cd Th best known of such types. is th~t cxamptc cxpansioDs callod after in which an arbitrary gencrally Fourier, periodic

02.

Thc

92.]

NORMAL

FUNCTIONS.

93

whose harmonies, periods of that arc submultiples of the given function. It is well known that th diniculty of thc question is confined to thc proof of tho of the expansion if this be assumcd, th dtermination ~OMtM~y of thc cooHjcieuts is casy Wlia.t 1 wtsh now to draw cnough. f~ttentio)). cases, physica.1 To
tmif'u)')a the gcncnd

function

is

rcsolved

into

a. sries

of

to

is, that

in

th

possibility considerations. our

this, of the

aud

au

immense may

cxptuisioli

of similar varicty from bc infcrred

fix
string

us consider the small ideas, let strutc)~ed bctwceu rixc<t points. that th wludc of motion, a. scries

vibrations We know it

of a from may each

thcory

wha-tever

motions, componcnt and capable function of tho time, rcpresuntcd of cxisting If we can discover thcsc normal by itscif. types, wc sh:dl bc in a position to rcprcscnt vibration thc most gnral an arbitrary to cach possible by combinmg thcm, assiguing and phase. amplitude
Aasuming

hc,

c:tn

bc

iuto aua.)ysG(t by a, harmonie

that

a motion tlie type

wo gct to dtermine

is Iiarmonic an equation

with of th

respect form

to time,

We assume

infer

that

is capable

gencral of rcprcseuta.tion

tlie most

which tho string position by a scrics of tlie form

can

which

case of Fourier's theorem. particular bc Jio dirHculty in proving th tlicorem in its most So far the

is

There general

would form. we ha.ve load the at ex-

has bcen supposed But uniform. string a variable only to mtrojucc or cven a single density, of th string, in ordcr any point to altcr compictely wliose pansion from th possibility may be inferred tlicory. wc stmil of It liave is to unnecessary furtlier examples such as dwc)l hre on this in th chaptcrs bars, membranes,

dy~amical

as subject, on the vibrations and connned

pa.rticular masses of air.

Systems,

94

VIBBATING

SYSTEMS

IN

GENERAL.

f93.

dtermination of the cncnicicnts to suit arbitra initial conditions hc may always enfected bv the fundarcadily mental ofthc normal property and Itmay functions, be convenicnt to sketch the process Iicre for systems like strings, bars, mem&c. in which branes, plates, thcre is only one dpendent variable considcred. ~tobe If tlie normal ~be and functions, th ~t, ~j, correspondtng co-ordinatcs,

93.

The

and

thc

correspond

is to problem with arbitrary mass

dctcnninc values of the of eicmcnt and dx, wc have from

so as

to

If p dx bc tite

(1)

But tiie cvery

the

products iutcgra.1

expression of tlie of tlie

for normal form

T in tcrms gcnGraI.zed

of~,

&c. cannot velocities, and

contain therefore

Hcnce of quations obtain

to

determine (4) by

7?, wc have pu, and intcgratc

only to over tlie

multiply system.

thc

first

Wo thus

Similarly,

93.]
The area, fact is just process or volume. The

CONJUGATE PROPERTY.
the saine whether tho lment

95
dx be a line, upon the is known

expressed conjugate property, that the functions are normal.

by (5), depends As soon as this

of a diSercntia.1 or othcrwise, we may by the solution quation infer the conjugate without further property proof, but th proth fundamental itself is most intimntely connected with perty variational of motion equation 04'. potential and p thc density displacement, clement of virtual dx, th equation 94. If be the cnergy of the velocitics of th dformation, (line, area, or volume) gives immediatety

is a symmctnca.1 function and 8~, of as may bc rca.dily from the expression for V in terms provcd of gencralizud In fa.ct if eo-or<U)ia.tcs. quation

lu

this

~F

now Suppose n. normal function with another normal

that

refera so tha.t function M,

to

tho

motion

~+?:~=0, then

to corresponding whilc 8~'is idontinod

as we arc cqudiy if wc suppose, c~tit!cd Agtuu, varies as M, fu)d 8~ as K~, we gct for th same quantlty

to do, that ~V,

from

which

th

conjugate by

presentcd A treatment of bars. good

rcspectively

folln-ws, if th motions property a.d M, have diffrent pcriods. of the two methods vibrations

rc-

of tlie connection example will be found in the chapter

of

on the transverse

9G

VIBRATINO

SYSTEMS

IN

GENERAL.

[95.

95. law on

Professor

Stokes' thon. p:~t~

lias drawn &f t!ie

attention f.ue

of a system not subject cKsp~ce~M?t<s titosc with which on tlie initial If velocities. forces, depend a velocity of any bo communicated to a system at rest, type and then after a small intcrvnl of time th opposite velocity ho communicated, tlie effoct in t)ie limit will be to start th system sponding to dcduce tbat without type. th but with a displacement of th correvelocity, We may rcadily from this that in order prove motion on initial from depending displacements on tlie initial it is only necessary to vclocities, respect to th time, express and which to replace th initial arbitrary velocitics by th

connecting the initial

to a vcry mut.iun which

gnral

dpend to fnction~l

depending diSerentiate with constants thosc

(or functions) which express th if ~) bc any

Thus,

initial corrcspouding displacements. normfti co-ordinatc the equation satisfying

of

which

thc rule,

first

term

may

bc

obtaincd

from.tlic

second

by

Stokes'

Dynamical y/t<'or;/ of Dt~'racft'on, Can~rtf~e

rraM. Vol. IX.

CHAPTER

V.

VIBRA.TING

SYSTEMS CONTINUED.

IN

GENERAL

the charactcr forces act upon a system, dissipative If two only of th of the motion is iu gnral more complicated. lincar transfunctions be finite, we may by a suitable and 7', and of th co-ordinatcs, formation rid our.setvcs of the products In the preceding we obtain t)te n<jrm:d types of motion. chapter <)C. WlfEN h:).vc conHidcrcd modifications thougb Electricity,
T

th ca.so of ~= will

0. 7'=0,

Tho

same or

whcn apply of impurtance in othcr scarcoly belong uf friction

with obvious theory but these ca.ses F=0,

of squares'; but th transformation them will not in general suit also the requirements to th form can thcn only he rcduccd gnral quation ~+~~+~+and to t!te simpler of frecdom, viz. not form applicable +~=~. to a system uc. &c.

Thc'prcscjice and to sums

such as Heat and parts of Physics, to our prsent subject. of will not interfuEC with the rduction for proper The of

(1), dcgrce (2).

of ono

~+~+cA=~i. Wc rcduce, l' and 97. may, though Y. There choosc whieli howcver, in Acousties tlie choicc

we shall pair of functions would almost always fall on

the which tlie theory then U~e most as T or when tarding R.

is, however, rduction of ait assumes

a not thrce

unimportant functions may simplicity. when j~is

class be

of cases

I)i

important V. The first motion

an exceptiona.1 are probably those case occurs part both

and effccted Under this bead form

of th same

thc

of cach

frequently, of th system to the mass

in books and

at any rate, is rcsistcd by a revelocity of th 7

force,

proportional

D8 part.
Iinear

VIURA.TINC Thc fun.) same

6YSTHMS

IN

GENERAL.

[97. J~ is form a

rduction is possible whcn cxccptioual of T'Mf'! K cr wbon 7' is itself oft~.r' ~mp

J.n any of tliese cases, t)io quations of motion are of thc samc form as for a. system of onc degrcc of frccdfnn, and tlie theory certiua which m{tke it wortLy of scparato possGsscs pccuHarities
cousidGra.tiou.

Thc ~nd

quations

of motion

aro

obta-incd

at

once

froin

in which For

thc the

co'ordinatcs froc vibrations

arc

scpa.rated. we Itavc oniy to put < = 0, &c., and

tlie solution

is of the form

and The

and whoc

are thc motion

initia! may

values thcreforc

of<~ n.nd < bo analysod into component to thc variation of but one in eacb of thse tlie vibration of a systcm with or less become only one

each of wltich motions, corresponds normal co-ordinate at a tinis. And modes is altogcther of libcrty. tlie similar After to tbc that a certain

dcgt'cc to the nmount cant, and

of dissipation, returns system with thc

thnc, grcatcr free vibrations to rest.

a.ceording insignifi-

sensibly frce

Simuttn.ncous1y of thon, pen<)encc tho quantitics tl\ dom, thc solution

but in pcrfcct indcvibrations, thcre may exist forccd vibrations on dcpending as lu. tlie case of ouc dc-groc of frecPrecisuly of

To obtain right-hn.ud and (~ vanish,

expression for (~ wc must cor)p!cte the initud member of (4), which makes th th terms given in (2) which rcprcscnt

n.Jd values th

to

th

of rcsidue

97.] at time

GENERALIZATION t of tho initia,! values

0F

YOUNG'S and If

THEOREM. there be no

99 friction,

th<Y-<!(.i(.'of'~i.u(~)rcducc;,I.j

of th normal co-ordinates indepcmlence leads to au interusting theorcin the relation of tho concerning motion to th initi.d subsequent disturhancc. For if tlie forces whicii act upon th system bc of such a clmracter ttiat' thcy do no work ou thc Jispiaeemunt indieatcd No such hy tlicn = 0. forces, eau produce on tho continucd, huwe~er long any cn'uct motion If it cxist, thcy cannot it; if' it do not cxist, destroy complte it. TI)c most important they cannot gcncratc of th application theorcm is wt~cn tlie forces apphcd to t!)G system act at a nodo of tlie uormi],! component tliat is, at a point which thc componcnt vibration in question does not tend to set in motion. Two extrme cases uf such forces an impulse, starting long that the systum turbed position. set in, and in tlie dennite charactcr conclusion rduction, 99. time.
So

98.

The

may bc specially tlie system i'rom is agai)i at rest

soon as

tho

(1) whcn tho force is rest, (2) \vhen it lias acted so under its influence in a disforce ceascs, natural vibrations noted, would in continue other for respects an intheir

absence infer

of friction that whatevcr

We

no component of thc type Tliis may be, thcy contain is limited to cases w!tcre T, F, F'admit of simultaneous of course tlie case of no friction. ineludmg The formutni quoted

in 97 are applicable to any Mnd of that wo have to deal only witli the force, but it will oftcu Itappen cnccts of impressed forces of tlie harmonie type, aud we may then more spcial formu)u3 applicable advantageous]yemp)oythe to such forces. In using normal we iiave first to calculate co-ordinates, tlie forces cl\, (1~, &c. to eacli period, aud thence corrcsponding deduce thc values of the co-orclinates titcmselves. If among tl)e natural periods without allowance for friction) (calculated there be any in with the pcriod nearly of an agreeing magnitude imprcsscd tlie force, vibrations will be abnormaHy corresponding componcnt indecd tlie force itself bo grcat)y large, ultless attenuatcd ni tlie preliminary
force of harmonie

rsolution. type

Suppose, and given

for
pcriod

example,
aets of at

that
a

a transverse
single point of

a stretched gnerai,

string. be excited,

normal modes Ail the not however in their own

vibration propcr

will,

in but

periods, 7-2

~0

VIBRATIN~

SYSTEMS

IN

(-.ENEKAI..

['~9.

in thc
v.hich

period
b:t~

of tbe
aL

Imprcssed
the p!j:nt

force
o!'

but

any

normal
wit! not

component,
bt; cxcited.

u. nojj

n,pp!iu.tuu~

Thc

of cach componcnt t)ms dpends on two tbings: magnitude of its notics with respect tn the point at which (1) on th situation thc force is appHed, and (2) on th denre of agrccmcnt betwccn its own proper and that of th force. It is import.fuit to period remembor that in

to a simp]u h:u')nomc force, thc syst.on respousc will vibra-tc in gcnera.) in ~ its modes, in pfn-tK'uhu' :dthong)i cases it ma.y somctimcs be snOicicnt to nttc-nd to only onc of thcm as bcirig of paramount importance. a.rc vo'y long oporating to th free pcriods of thc systcm, :n] cqui!ibriumthcory rc)~tivc!y is sometimes but in such tlie solution n. ca.sc could ad~uate, Le fuund more casily without thc use of th nonnn) gcnc!Lt!y co-ordina.tcs. BcrnoulH'.s :Lnd class, Dicory of thc Tides is of this on thc assumption that thc frcc pcriods of' thc masses of proceeds watcr found on tbe globe are s!n:d) rdativdy <.o thc pcriods of thc pcriods forces, operative Icftoutofaccount. in whicli case thc incrtia, of fact of thc this water As a matter supuosition we arc conseqnc'ntiy bc might is on]y 100. When tho of tho forces

and pa.rtialty and vcry rougidy applicable, still in tbe dark on many important to thc tides. points relating Thc principal forces have a scmi-diurnal pcriod, whicb is not sufnto tbe natural to a)!o\v concerned, ciently long in relation pcriods of th Incrtia of Ibc water buing ncgiccted. of But if th rotation the cartb bad bccn much slower, of the tbe theory cquilibrium tides migbt !)ave bccn
cquDibrium

adc(ptatc. is sometimes w])en thc uscfuL t!)cory force is sumcicntiy in compar!son long but not so in thc of a System, poriods It will In thc bc sufDcient quation + ?t~ = varies as cos Theti to ta){c thc case

A con'cctcd

of tbe imprcsscd pcriod witb most of the natund case \vherc of onc or two tucre is no friction.

of thom.

f?~ + c~) = <~, suppose tbat t)ic imprcssed force

or

100.]
S)!ppo'.<)o\v

EQU1LIHHIUM
<'))f~thi~t'u'c'sj~stif!;(,bk\f'xcfpti)(:p''ct

THHORV.

101

uf thu to th the truc

sin~te rcsult and

normal uf thc
thc

co-ontinatc
cquitibrium

~),.

Wc

)):LVC tho)

only

to :uid betwcou

tliere

;),snut)ied

the diircrcncc thcary, v~luc uf (& viz,

Thc ftu'ccf!

other
vibrations

extrme bc of th in

case cxtrcmciy potential

ought

aiso

to

bo

noticcd.

If

thc

rn.pid, enei-gy with ~ivcs

iodupendunt of ne~cctin~ ?; iu comparisofi

comparison with wlucti

becoino they may ne:u')y of the system. Instout wc ]i:wc thcu to ncg!cet

If tbere is not Lhc

Le onu

or

two

co-ordinatcs

to w)iic)i thc result,

this with

trcatnicnt on con'cet.ious

wc i~pplic~bic, th:).t hypothu.sis particular Before

fur thse 101.

may suppicniott is !t)t.ogct!tcr co-ordinates.

calcuintcd

nc~tigibic,

passing

of .Systems snbjcct Home pcculiaritics

startcd by

to

on to t))c ~encml theory of thc it may bc well to dissipation, free at vibrations point. at of onntinuons On thc any a single con6guration

vibrations point out

uf thc applicJ ~8, tbe

Systems,

a force of

aud jnincd

notations bv

suppositions time is detcr-

Suppose at thc poijtt tion

itnprcssed

now that (?.

forces

System T))C value of thc

tho

is held donc

n.t rest

is detjcrmincd upon
disptaecmcnt

by a, force applied by th considemth System hy tlie S~=S6

tha,t is

<I\8<

reprcsGnbs
d'n'ing

work

a hypothetical

that

102 If twe th a.t any

VIBRATINQ system sub.uu't hc let

SYSTEMS go from t:nie <

IN this

GENERAL. Cuiif!?ura.tion

[101. at<=0,vc

and a.t the point

neither

nor diverges, converges, wltb t)~ converges Again, suppose from th configuration that thc

with

r.

Th

series

for

~thercfore

system

is

started case

by

an

impulse

of equilibrium.

In this

initially

Dus

gives

shcwing is more

that slowly

in tha.u

Uns

case

the

series case.

converges

with

n,

that

in thc

prcvious

101.]

SPECIAL

INITIAL

CONDITIONS.

103

thah tho value is of mny br* nbsnrvcd with respect to 2-' aud tliat tho disptaccsymmctrical proving ncnt n.t time t for th point 7-* when the force or impulse is apSinne it would bo at () if tlie force or impulse as pitpd at < is the both h:td bueu reciprocal As a thit'jd as arca, dcfur)ned {),pplicd theorcm,
case

In

M3.

it

at

of a vcry general example which we shall consicler at !eugt)i pt'csc!itly. wc th body to start from rcst may supposa force M)!bn~y Hud rcadily f~M~M~c~, over its lcn~t.1),

-P.

This

is an

by or vuluinc.

\Ve

than whcn thc force will hc more convergent for is conccntrated i)i a siugtc point. In exactly tlie sa.mc w~y wc may trcat th case of a continuous motion is Eubjcct to dissipn.tion, body whonc pruvidod tliat th tlirce futictions bc simulta.ncousiy 2~ J~ reducible, but it is not necessary to write dowu tlie formuJ. 102. If th three mccha.nica.I T, -F' and V of any tlie natural vibrations reducibic, arc moru in tlicir complica.tcd dissipa.tion this class form is small, the of casusbcsidcs mctttod being introduction to 2' and V arcc functions

The

series

be not system simultancousiy bcen (as has aiready observed) charactcr. th Whcn, lowever, of of tlie rduction sonc more is still importancu gcncrat as sums

and usofnl; in Itscif will We theory. of squares

suppose

a good thcu. that

cxprcsscd

Thc quations

of motion

a.re accord!ng]y

in ]f

which tlicrc

the were

coefficients no friction,

&
ttic

&c. arc
abovc systcm

to

be
of

trcatcd
c(}uations

as small.
wuuld

JU4 bc will satisfied t,!ic

VIBRATING

SYSTEMS

IN

GENERAL. to

F 10 3.

while

onu co-ordinate by supposing other co-ordinatc-s vanish. In

the

vf~' suitahly, actual case thero value of any ot)icr

be a corrusponding sohttion in which the will bu small rclativcty to 6, co-ordi!mtc Hence, bceomes, if
wc omit tc'rm.s of

the

second

ordc-r,

thc

?'

equation

from were

which no

wc

infcr duc

tliat to friction

varies

c!)angc

:LS if n-pproxim~tuly in thc type of vibration.

tliere If (A

v:u'yase'wcubt:utitodct(;rmiuc~

Thc is small From a-M tho

roots

of

this

in eomparison thc

arc comptux, quation with tlie imaginary part. if wc introduce as e" we gut tlie

but

tho

real

part

quation, vary

supposition

that

co-ordinates

dctcrmines cquatinn thc altcrcd approxiniatdy type of vibration. Sincc thc chief part la ima~hary, wo sco of tliat thc co-ordinn.tcs arc in the sa.me phMc, a.pproxi!natc!y ~<~ </tC6< ~j~Me f~y'e~ ( ~MM?'/er per~o~ /?'o!~ </<e ~aM Hcnec wttcn thc function F docs not rcduee q/' to a sum of squares, thc chamctcr of thc modes of vibration c]cmentary is ic.ss simp)u th~n othei-wisc, aud thc Y~rious of tlie System parts arc no longer in thu samc phase. simuttanconsly abovu that, w)tC!i titc friction provcd is small, the value of y?, may bc calodatcd without aUowancc for approximatuty thc change uf tyj)e but hy means of (6) we a still may obtain closur approximation, in winch thu squares of th small quaritities are i-ctahied. Thu ?- quation (3) givea We

This

~e!u:uimg' thc con-uctiun

titc r:T.te uf

tenus part ofthc lias no uifuct un (lej)cnds.

mdudcdundcrS Litc ruai p.n-t

uf

bcingrcit), oa w}dch

dceay

103.]

SMALL

DISSIPA

TIVE

FORCES.

105

103. cquations If wc !ia.vc

Wc

now

returu

to

ihb

cunsidratioti

uf

thc

g.'n<t..J

of 84. &c. bc th co-ordinates an<I &c. tlie forces,

For tcnninant

thc t

free

vibt'atiuus

~F,,

&c.

va.msh.

If

\7 bc thc

dc-

thc result

of elimiuating

from

(1) aU t!tc co-ordinates V~=0.

but

onc, (4).

is

S~ncc cquahon piurs, compJutu Lut

\7 nnw co))taius = 0 no jouter cot)ti).in intgra!

of' D, t!)G 2?~ roots of tho powcrs uccur in equal :uht ncgativc positive !). ruai as wu!! as an imagmary TIte p~rt. stil bc writtun +. uc. solution such
&c.,

odd

may ]n)w<;vcr = ~c~ + J'g~~

+ Z?e~ + 7/e' are

(5), Corruas

where

thc

Mpoiding

pairs uf cunjug-u.tc to cach rot, thcro

~=~~ ~=~

roots

is a. particular
~,=~

in

which

tions

arbitrary. In t!te prsent case ]iowcvur contains odd powers of Z)) (wlicre thse ratios aru not in gcncral th variations i'c:d, and therefore oi not synchronous in If thcco-oi'din:Ltes'&c.:u'c phase. we put &c., wc sec tha.t none of th /~=a,+t/3,, ~=a,-t/3~, a can bc positive, since in that quantifies case thc energy of thc motion would Incrcase with the time, as we know it cannot do. Enoug)i tions of a
bas System now in

?Yt~'os j'l, of motion, and oniy

thc

arc thc absohjtc

determined value ronains

by thc

equa-

beeti

said

on thc

illustration that it general. Any will bc anorded of the case of two may rcqnirc hy t!)c discussion of frccdom, of strings and uthur 112, and by the vibrations dugrees bodics with whicli \vc shaU soon be occnpicd. We rsume spcial ti)e nature quations (1) with ~m~to; of forced thc view of
invcstigating

snbjcct further

of the

froc

vibra-

further

tbc

10G

VIBRATING

SYSTEMS

IN

GENERAL.

[104.

104. ot-dinatcs minor

In but

or~r

to

cnmu.atc oper~tc

from ou
tilcm

thc
in

crjnnH~ns
succession

~!L t'~
with

.)the

onc (~), dctenuinimts

and

a.dd the

results We

co-ordinates. system

togcthcr; ttius obtain

and

in

IH

manncr

for of thc

thc

othcr

as the

cquivalcut

urigina.!

of quations

in

which

the of the forces condition that

dincrentiations cquaHty

of

cognition Thc to tlie motion

subsistmg &c. arc

~7 are to botwecn any whatcver, so grcat

be e,

made and e

without

re-

of

not

become squa.res sensible. quantities t!ie case, the forces opcrating If, as is ofteu he !nade up of two one constant with to timc, and tlie other periodic, parts, respect it is convenicnt to separatc in hn~ginn.tion tlic two classes of cncets T!ie effect duc to tlie constant forces is exactly produced. the same as if they acted and is found alonc, by th solution of a statical It will therefore bc sufficicnt problem. gcneraHy to suppose th forces pcriodic, tlie effects of any constant forces, such as gravity, to altcr t!tG configuration about being mcrcly which tlie vibrations arc exccutcd. Wo may thus without proper rcat loss of gcnera]ity confine ourscives to perlodic, any and therefore thcorcm to harmonie forces. by Fourlor's thereforc assume as expressions for ~P,, &c. circular might functions of th tune but, as we sliidi have occasion frquent to recognise in th course of this work, it is usualty more convenicnt to employ an such as imaginary exponential function, where~Is a constant which ~'c' When th may bc complex. solution is obtained, its corrcsponding symbolical real and Imaginary tiie real parts may and Imaginary be separated,
parts of

tlie

producing of th small

of course, subject, a displacement or

Wc

and thc

belong respectlvc!y data. In thia way

to tlie

104.]
~n!)Jy-t['! gam.s consider~biy

FORCED

VIBRATIONS.
iu

107
~s ditfcrpntmtton!!

brevity,

Inn.smuch

anfl

altrations

th compicx We therefore

a.rc expresscd phnsc cofHcicnt without chang-ing write

of

by thc

mcrcly modifying form of th functiou.

Th fonctions thelaw

minor

dctcrmmanta Bymbo!

of the 7), and

type
re

arc on

rational &c.

intgra.! to

of the

operatc

according

where bolical solutious

&c. are arc

certain

complex

constants.

Aud

thc

sym-

wherc

(~)

dnotes f!t'st the

the

rcsult

of substituting

for D in

of a System from friction. exempt ~7 coefHcieuts and its (liHereutial arc titen c~M functions of D, so that of ~7 (~) is rca!. Tbrowiug a.way th imaginary part for ~t~ &c. wc hve thc solution, writing ~' Considcr case

If than may

we one

suppose goiera.Uzcd

tliat

the

forces liave

&c. ail

component)

(in thc case of more thc same pliasc, they

be cxpressod

by

and

then,

as with

in phase

casily tlic forces

is

sccn,

th

co-ordiuatcs

themsolvcs

agrcc

Th on the bc the (ip)

amplitudes

of

tlie

vibrations Now,

of \7(?'~). magnitude same as one of those = 0, a.ud tlie

dpend if thc

among

to bctonging becomcs iniiiiite. amplitude

period tlie frce

tlungs of th forces vibrations, This is, of

othcr

J08

VIHRATINC:

SYSTEMS

IN

GENERAL.

[104.

course, just thc casu iu Ct'nHtdct'f~tiun of friction, exempt.

which from

it wltich

js essential no natural

to

introduce

the

system

is ruaHy

bc friction, but it may bc dividcd ~7 (ip) is eompiex into two pa.rtsonc rca.1 a)id thc otho' puruly ofw!iic)i Itn~ginary, tlie latter dpends on the friction. if wc put Thus, entirely V (~)=~ \7.~ are eveu our J,=Vt\e~ functions
solution takes

If

thcrc

(~) of thc

+~ and form

V.(~). therefore real. If as

(7), 1 bcforc

on thc friction but dpends entircly ~(~) it is not truc, on the othur hand, th:).t 7,(~) is cxa.ctiy thc s~me, as ift)tcre had been no friction. this is approximatciy Howcver, the ca~e, if the friction ho sma!]; bccausc any pn-rt cf ~(~), which on thc first power of th coe~cients of friction, is nocesdpends sari)y period imaginary. of thc force W!)cncvcr and tliat tan there of onc ce is n, coincidence of th frcc :).u<~thcrcforc between tho \7~;)) vibratious,

Wc

ha.ve said

that

va.Dishcs,

f~nd we ha.ve

y =

indicating friction. On thc so also With

samo

a vibration

of large ofsmai

amplitude, is

only

Hmitcd

by

the

hypothesis

friction,

in general

is T, except in case of approximatc (.-qnality certain the motion bas nearly exceptions, thereforc, with tlie force that excites it. (or opposite) phase a. force

smaU, and of pcriods. th

term acts on a system, expressed by a harmonie the resulting motion is everywttcrc and rcta.ins tlie harmonie, that th squares of the displaceoriginal period, providcd always ments and velocitics This important may bc neg]cctc<1. principle wa~ cnuuciatcd and a.pplicd of by Laphicc by him to the theory

Wlicn

104.] J
the tidus. Hcrsche!, If th of vibration
from each

INEXORABLE MOTIONR.
Its.grcftt to witom force

lOf)

was atso rccogt)!scd gcncratlty hy Sir John wo owc a formai domonstr~tion of its truth*. function system of th of the time, thc types diffrent

bc not a. harmonie parts

from

'ni dtfferent
other and

of the that

are in. gnral force. Thc

harmonie

fonctions

are

thus

th

wldcii, as was changed, that rcason for anticipating 105. We now tnrn

on!y oncs winch rcmn.rked in thc thcy

their prserve type nnis a strong Introduction, to simple toncs. correspond diffrent as hitherto, Idnd given of forccd inexora-

to

a. somewhat forces

instcad where, vibration, ble wo~t'o~s are prescribed. If

ofgiven

t)ta,t the co-ordinates we suppose while th forces of the thno, ftinctions thc quations vanish,
selvcs Into two groups, viz.

arc givcn of rcmaiuing of motion divide thc types thein-

In cach tcrms cH'ect this

of the

~?'

are known as know)i gronp are after

cxphcit furccs a.cti)ig on thcrcfore su~icicnt

quations fnnctions

of thc of thc the

latter time,

thc first r group, and hve thc sa.tne

Thc Ct~uations of system. to (tt.'t.crminc thc uuknowu

quautities; ta,m tlie

group. the two classes 10G.

slinpic of sorne

thc forces ucccss:u'y to maLinwhic)i, If rct~uircd, motion from thc rirst may bc Jetermiucd pi'cscribcd tit~t thcrc is no esscutial diffrence betwciCti It is obvions of prohtcms motion of forecd
vibnitions.

The

harmonie
of thc

of a systcm dcvold vibrations in consquence

posscsscs a

of friction

and

of prescribed
paraUel

cxccuting variations
to thosc

coordinat.es,

pcculiarity

considcred

in 74, 7~.
= J~ cos

Let
= -j eos ~<, &c.

in whicit

thc
7~)cye.

quantitics ~<~ro~. art.

~l,. 823. AJHO

arc

regarded
<

as givcn,
~"<rf'))fMy, fino.

whi)c

thc

0)<f/t'))f.<

110

VIDRATING

SYSTEMS

1~

GENERAL.

[lOG.

l' and ~~T

from V, 82,
2(y+ =~(~+~,)~/+.+(~+~)j~+.,

tlie

expressions

for

+~(~u)~+.+(~J~~+.jcos2~

from dition

whic)i th.t

we

see

that variable

tlie

A', tlie Hc,J~+.

quations part of y+

of motion r, ~i,ich

express

tlie

conto

is proportional

shall ~r.,
vjbratiug

bc ~I,

station~-y Lcb under tlie

+(~)~~+,~ in v~uc, for variations of tlie claautitics bc ttie value n~turat to thc System of~ wlieu restraitit dcHned Ly tlic ratios

Krom if the

this

we sec

prc.scr.bed the system uuder

be ecrtainfy less than be grcatcr pcriod than any of tllose the partial constraint rcprescuted by ~t.J,

that

if

tbat natural

is, to

is necessarlly aud tl~e .tationary positive, val~e-t!~re can ouc-~ an absolute minin~un. For a similar rea.so~ if the prs nbed ponod be less tiuL.i any of tliose natural te thc pa,-tia)iy constraincd is an absolute System ~xhu~ but a~braica!Iy, an absoluto arit)nuctieat!y rnini.num. But whcu lies witbin thc of range vaines possible of~, or ngative, n.ay bc positive and th actual value is not th greatest or least possible, Wi~enever a natura! vibration is cor~sistent with the hnposed that conditions will ue thc vibration assumed. Tj.e y.,l.Uc part of ?'+ zero. For convenience two grt cl~es of treatment <.f forced vibrations .ve hve considered t),e apart but hc c tl.e lapse o

then bc but

vibrations; is, of course, noth.ng to prcvcnt their coexistence. After of a .sumc~nt interval of time, the frce vibrations ahvays howcver appcar, small thc friction The case of abso.nay be. uo fnchon is lutely idca!. purc!y Ti.crc is onc caution, ~Lich lowever, may not bc supGrrIuou. in respect to thc case whcrc givon ~Jare forcj~

and f~

106.]
System.
givou.

RECIPROCAL THEOREM.
Suppose,
Thf'u ~t);

111
arc
or .non-cxistcnco

as before,
.frc't) \'ibt'it.{.iui')~

that

the
wh'

co-ordinates
'~L-n<x'

is a matter mnst the order

of indiffurenco to

Le understood co-ordinates

to preveut their must bc iutroduced; so that from one point of view th motion in as forced. But tlie applied question forces are may be regarded of the natu.rc of a constraint; and their ct'ect is the same mercly as a limitation on the frecdom of thc motion. 107. Very rcmurkable reciprocal forces aud motions of different types, extensions only pose zero, For that we
relations exist

so fur as th forced motion is concerned, be such as th system is capable of, when are not aKo~ <o MtryJ~'o~?~ zero. In forces of thc corresponding varymg, types

between be

tlio

as regarded of th cerrespondi.ng theorems for systems in winch T bas to be considered If we sup( 72 and 77, 78). ail thc component and ~F arc forces, except twoF obtain from 104, may

which

We whcn~ va.nisitc.'i.

now considcr vanishes, Ii"~K=0,

two Mtd

cases secondiy

of motion (with

for the same da.shcd

first system whcn Ictters)

coefficients ~7 and its dticrentia! quations functions of tlie symbol and sincc m Intgral D; and thercfurc ~r. = ~.r' V is a. symmetrical detcrminaut,

In thse

arc rational cvury case

Hcucc ordinatc is retatcd a.Lone. lu frcc

wo

sec

that

is rclatcd to tlic force

tlie cosystem, to it in the same way as the co-ordinate whcn this latter foi-ce is supposed to act motion hre may be at th cxistin~ thcre

if a. force

act

on

the

addition

to

thc

vibrations

dpendent

contcmplated, on a. disturbance ah-eady

112

VIBRATINO

SYSTEMS

IN

GENERAL.

[107.

moment included iorc~s

subsquent in ~F; but \t'hicii acLed

to thse

which

ai!

new

sources are

of disturbance tho th e~-ct

are of

vibrations

thcmsc)ves sinail

dissipation be, tliere must of time after which free vibradie out, and beyond winch it is unnecessary to go in taking account of the forces wbich hve acted on a If therefore system. we include undcr forces of sumcicnt there are no reinotcness, vibrations to be considered, independcnt and in this way tlie theorem to cases which may be cxtendcd would not at first sigbt to corne within its scope. appeMfor example, that the Suppose, is at rcst in its position systcm of equilibrium, and then begins to bc acted on by a force of the first in type, graduatty Increasing from zero to a finite value magnitude at which point it ceases to incrca.sc. If now at a given epoc)i of timo the force be sudand reinain xcro c-vur afto-wards, dcn!y dcstroyed frc vibrations of tI)G systcm wIU set in, and continue until destroyed by friction. At any tirne t snbsc()ucnt to th given cpoch, tlie co-ordinate has a vatnc dpendent upun t proportional to T))c tiicorem allows us to assert that this value bears the same rcJation to at t))c same mnmcnt hve borne as~outd if thc original to~ cause of the vibrations bad been a force of th second type infrom xcro to creasing g.-aduai)y and thcn suddenly vanishing at th given cpoch of timc. Wc ))avc ah-cady had an of example tins in 101, and a like result obtains wiien th cause of th may tions of the Impulse, or, as in the problem a variable force of finite pianofortc-string, short duration. though ln tljesc app)ications of our theorem we obtain results to rclating h'ee vibrations, considercd as t]jc residual effect of forces whosc actual opration may ]tavc becn long bcforc. ~08. harmonie, J,e~ forces results be may In an important a)td of th samc wherc J, in thc same then and
e)ass pcriod. of cases We th may forces rcprnscnt and them are by

previousiy. be an interval

However

origin:).!

disturbance

is nn

J~

t-hase be written

may be assumcd at the moments

to bc ?compared.

If tbc Tito

108.] SIncc tlie ratio

RECIPROCAL Y~

THEOREM-. same

11~ is

is by hypothesis th real, true the of th ratio which tha.t signifies are iu tlie same represcnted by those symbols phase. to rcal quantifies wu tU~y state the thcorem thus If a force <<? ~to~t'o~ ~ro~t;cc s~/s<e~ A~ cos pt, c(c<t'H~ o~ ~e = 0A, cos (pt ~cn wt'~ (t force e) </tC ?~o<K)?t '= ~A~' cos (pt c). Le no friction, '=~. are of e will Le zero. But not it must be remembcrcd =

motions Passing

to ~n'6 rise = A~ cos pt

If thcre If

thcn J, = thc forces ~F, and than thc co-ordinates example

p!accmcnt.

that

nccessarily

may

represent

corresponding a linea-r and

any more comparable, for one of wluch types, disanother an angula.r

Thc before not

reciprocal

theorem

to thse proceeding give a knowlcdgc of dterminants. requiring and and 103, corrcspuuding give disptacctnents,

may bc we will

statcd

in

sever~l

ways,

but

another

investigation,

If of forces

be the

two

sets

quations

of

motion,

of a symboUc vary as e' the cfFect such an e~ ou any of thc quantttics to is mcrcly opcrator that found multiply quanti ty Ly thc constant Ly substituting this substitution ?'/) for j9 iu Supposing juade, and havicg = e~, we to tlie rclationa regard e~ may write New,

if

ai

the

forces

wltich ~9. will two zero.

is tlie lu

expression thc

oi'tlie

reciprocal that that take we the as

ruiatiuii. arc forces the first abuut of and to ai makc types secoud) it In:t are

applications

snpposcd throughout (whicli wc mn.y as well Thus

be

114
Th dinerent

VIBRATING

SYSTEMS

IN

GENERAL.

[109.
in thrce

con'icquences In ways.

of this tlie first ~=0,

equation we suppose ~/=0, ~=~'

may that

be cxhibited

whcnco as before, that the relation shewing, to in th first case when ~==0 is tlie same as th relation of to in tlie second = 0, tlie case, when of relationship' exidentity to phase as well as amplitude. tending A fe\v examples the comprehension of a law, may promote whose extrme is not unilkely to cuuvey an impression generality of vagnencss. of a horizontal Q bc two points bar supported in one end clamped any manner and tlie other (c.g. with a frce), harmonie transverse force applied givcn at P will give at any moment the same vertical dencction at Q as would have been found at had the force acted at (~. If theorcm same Or we take will ?'o~~o~ if one angular l'un :-A displacements, harmonie given couple at P will give as the couple at would givc at P. the ot))er instcn.d of lincar the the If .P and ~(2), of

at

acting th force at Q, as th maximum force at <~ would do in acting tlie maximum through at Q due to the couple displacemcnt at P. In this case th statement is more compHcatcd, as the forces, being of different cannot be taken kinds, equa!. period of th forces to be excessivoly tbe momentary of the system tends to concide position that in which it would be mamtained at rest by th then and tlie forces, becomes equilibrium theory applicable. theorem thcn reduces to th statical one proved in 72. As air, a second example, whol]y, suppose or partly, that connncd in and either a space by sotid occupied boundaries, If we suppose thc long, with acting Our

bc Jincar and dispJacemcnt rcsult thu.s Suppose for may be stated harmonie acts at .P, and for th couple force of the same and phase acts pcriod at Q in t!tc first case bas displacement same as the phase rotatory displacement and tbe amplitudes of th two dplacements th maximum at P would do the couple over th maximum rotation at P due to

th angular, thc first case that a second that a vertical at at at are same Q, thon cvcry in th moment th linear tho

second, so related that in

work

by

109.J thcre arc two

APPLICATIONS.

115

and whosc centres sp)tercs have one denre of freeclom. Thoi a. periodic force acting on ~4 will produco tho same motion in j?, as if the parts werc and intcrdtangcd thi.s, wliatevcr mcmbra.ncs, forks on rsonance strings, or cases, other bodie.s capable of bcing set into vibration, in may be present their neighbourhood. dnote two points Or, if A and of a solid etastic body of any shape, a force paraUcl to at acting A, will producc the same jnotion of the point to Oras an cquaL force parallcl to Oy at would pn.ra!Iel in the acting producc point ~1, pM'a!!cl to ~J~. Le two points aga,in, lot A a)nl of a space occupied by between which air, arc situatcd obstacles of any kind. Thcu a sound originatingat A is perccived at B with the samc intensity as that witit which an cqual sound at jS would be peroriginating ccived at ~i/ Thc obstacle, for instance, consist of a rigid might wall picrcecl witli one or' more holcs. This example corresponds to the optical if that of renectin~ law by any combination or re. fracting can also arc surfaces bc seen one point from thc be seen from a second, the second first. Fn Acoustics the sound shadows in consquence of the not insignificant in comparison with thc dimensions of rcciprocal be of taken wire
relation is of

Or

can

usually only partial value of th wave-Iength obstacles and tlie ordinary interest. A further be two bourifood example may circuits of insulated combination with

considerable Lct their there

from

/1 and

electricity. B, and in or solid

any in communication force in th circuit as

inR

wu'c-eircuits

neighconductors

condensers. A will give rise in il if the

A periodie to th same

electro-motive currcnt in force opcratet)

would Our last

be

excited

cicctro-motivc

will bc takcn from th theory example of conduction and radiation of heat, Ncwtou's la,w of cooling assuvned being as a basis. Th temprature at any point ~t of a and conducting due to a steady radiating system source of hcat (or harmonie) at is th same as th temprature a.t due to an equal source at if at any time tlie source at B be Moreover, th removed whole subsequent of temprature course at A will be the same as it would be at B if th parts of.D and A were interchanged. 1 Helmhoitz, Tho Bonnes must be Rnch as iu the absence of C'r<~< Bd. Lvn. obstacles would diffuse thernselyos eq~Iy in ~11directions. 82

n~
HO. fu'rived The

VIBRATINH

SYSTEMS

IN

CHNKRAL.

[110.
theorcm is

second

way

of

stating

the

reciproc~

at by t~king

in (1) of 1()9,

that shewing is the samc wheit ~=0. in

a

thc relation as the

of

to of

ni th to

nrst

relation

case, wlien second in tlic

= 0, case, held at and and

Tlius rcst wliilc

tite
givcu

cxampio
vibration

of the
is

rod,
imposed

if thc

point in

be

applicd),

th

reaction bc

at jP is thc at if that imposed electric were

upon same hotli

(by a force amplitude beld at rest

thcrc

as it would phase thc givcn vibration So if J- a.nd of any uumber in condensers,

were point upon 7~. circuits in

bc two

th

closed of othcrs, whether C, D, in ~1 by and a givcn periodic enrrcnt bo cxcitod cicetro-motive cicctro-motive Induccd th force, th necessary of ~1 as it wou!d be In ~t, if thc parts force ii) is thc saine and TItC of 5 109. wcre tinrd Intcrchangcd. form of
statemcnt is obtaincd by putting

ncighbourhood or terminating

in

(1)

proving abne, case,

that is tlie

th

ratio

of

to

in tho

first

case,

\vhpn in tlie

nets second

to of tlie ratio of ngative as to kecp when tho forces arc so rclatcd if force to the th P of the point acts at Q, tho rcaction force at at Q as thc P, if th rod be held at P bears

cqual at tho rest samc

to zero. while numeria

Thus periodic cal ratio to th a force 111.

displa.ccmcut at .P. applied Thc

at disptaccment rod wcre causcd

bcar Q would to vibrate by

for ait Systems theorem bas been proved reciprocal forces can be represented in which the frictional F, by tlie function of a further and an important but it is susceptible generaHzation. F for of the function We have indeed th existence proved a large proportional class of cases whcrc or th motion is resisted but to thc absotut.u relative velocities, by forcf's theru arc

11L]
oth~r sources of

TWO
dissipation

DECREESOF
not to

FREDOM.
be to or in to brought include radiation thse the cases velocitics under this

117
hcad, for exemple, th of hcat. Now arc not for ~M or displacearc exdisthe a~ sonse

whose

effects

dissipation f).tt))ough ~)~6'~e ments, pressible placement

it is eqnally important due to the con(htction it bc truc that the forces in a constant any actual ratio case of in

~huns yct

ncccssarity

periodic, by a sum (absolute of th part not

motion pcriodic (T) tliey and wliatevur tlicir tttcrcforc, phase, of two tonns, one proportional to th or relative) and tho other proportional to of the system for ail fmiction In aneetcd. motions ~exists fact If th whatever, in thc it &br only

vulocity bo thc same, motions u/e~

coemcicnts

ncccs.sarity T, the

for our prsent requirud purposc. with motions of pcriod T titat a matter of indiiTcruncc p):un!y are dpendent T or not. upon pct-haps
forces.

is exclusively t)ie Dtcorcm is concerncd, it is whct)icr tlie fonctions F Y; Thus tho theoi-em is cxtendcd, wtiole ricld of dissipative

since

sufliciently

gnera)

to covo.- tho

It is important to remember Hiat the Prnicipio of Reciprocity is ilmited to systems which vibratc about a configuration of e~tand is therefore ~M~)t, not to bo apptied wititont reservation to such a problem as tliat presented of sonornus by thc transmission wavcs tbe atmosphre wi)C)i clisturbed through Thc by wind. vibi-iLtions must also bc of such a charaeter that tlie square of the motion can bo ncglectcd our dmonstraot))crwise througitout; tion wou!d not hotd Other apparent good. on exceptions dpend a misunderstanding of thc principle itsclf, Carc mnst be takcn to observe a propcr between the forces and discorrcspoudcnce the ruie that th action placements, of th force over tho being is to represent wo~ ~ne. disphccmcnt T)ms co!~)~ correspond to to )'oMw:s, inercmcnts of ~re~M?'M ~'o~trne, aud so on. 112. In III. we considered Chapter onc degrec of frccdom. TIie will he devoted to sonic detaits arc two. two co-ordinatcs, t)ic expressions for 2' th vibrations of the of a

with system sent Chapter degi'ccs If and

remainder of the case

prwhcre the

of freedom and

F are

the y dnote of the form

118

VIBRATING

SYSTEMS

IN

GENERAL.

[112.

so that,

m thc

absence

of friction,

the

cquations

of motion

arc

Thc

constants

Since
juust

l' and
be satis~icd

L, J)f, -~V; ~1, 7~, C, arc not entu'ely arbitr~u'y. F arc essuutially thc foHowing positive, mequa)itics

Zy>

~1C>~

.((!).
on the

be positive. Moreovcr, L, N, ~t, C must thcmsdvcs tu examine th ciTuct of tliese restrictions We procecd roots of (5). Iti positive. cocMcicnt th
For

first of

t!)e tin'ec coefficients in the equation place the first and third, this is obvions from (G).

are Tlie

winch, as is sccn Wu concludc J/7It rem:t.Ins (.Uti<j)t to bc

tive

from thttt

(6), ~/ZV~6~ tlie vulucs of

is ncccssarily than grcatcr if rca!, arc both ngative. m ihct quantity rcal. be Dm not eoinga-

to provc that tlie roots are satisHcd is that thc i'olluwing

~A~lC-Jt/~ is positive. This is titc au~ytica! uf are proof th:tt thc vaincs hoth rca! and ngative I~vc bccn anticipatud a fact tlu~t might -iUtout from titc pltysica.1 constitution of :uiy an:dysis thc systum,
satisncd, whosc vibrations thoy serve tu express.

whieh

shows

that

thc

condition

is

since

113.]

ROOTS

0F

DETERMINANTAL

EQUATION.

119

Thc two

values

of

are different,

uniess

&o~

Th

common

spherical

pcndulum

is an example

of this

case.

of a suitable force F the co-ordinatc By mcans may be preT]to systcm thon loses one dugrcc veuted from varying. of frcconc is i)i <!o:n, and thc purnjd eon'cspouding to thc rcmaining beforc thu introducof thosc possible diUbrent from cither general tion. titus e~ tliat tlie types Suppose iu tuni the variation preventing of 3~. Tlien (L~ bcuig that obtained bc put Zy(~) wltich modiatc tion tities /(~) wc sec tbttt deercasca whcn shews in will provc at once value that and tliat from into are thc roots of the motions obttUtied rcspectively by and x are of of the quation

+ A) (~V~ -t. C) = 0, (4') by supprcssing the form (~ ncitlicr a.ud -~) = (~+ of tlie roots A little B)' and R Hclice

(4) may itscif

(8),

of X" can be interfui'thcr examina-

betwoeu the othcr

onc of the routs

is grciLtcr than hoth the quan.loss tha!i both. For if wc put (~ smati, to f f ~) (.V~ + ~)=, (J~J~); sign and when bccomcs

= L~(\' is vcry

/)

is positive dingos is thcreforc

there a root; imd :t.)so ncgativc. bctwccn aud ce. Wc conchtdu tha.t thc by sintii:).)' rcasoning t)ic systcm to tbe two kinds of contones obt:t.iued by subjecting In pitc)i bctwecn in qnustio)i ar bot)) intunncdia.te tbc stnuut of vibrations tlic system. lu p:ii'ticu!a.r tonos giveu by t)~G nntuml cases /t may bc cqua.1, and then

(id~cbr&icaDy) Bctwcoi 0 a.nd

mn.y bc gcnerulixcd. proposition wLidi Icuvcs thc System still in posscssiol dom ~s thc impositiou may be rcganicd t)ic co-ordiuates, such as bctwccu This

~h?y ~~nd of constr~mt of onc (tcgrcc of frceof n, fot'ccd relation

120 Now tiult

VIBRATING if Cta? + .y,~o tlic a.nd

tfd!cn

SYSTEMS any
~s

IN

GENERAL.

[112.

r.f .'s rmd

tha.t provcs introduction thosc ibrmcd. whcn period. If tlie two

single pcriod of tho constmint, in which tlic natural

Hncar funchomogeneous n~w t.ijo eftinc argument vnrinUcp~ to thc systc'm after t)io possible ja intermc(U:Ltc in va.!nc bctweoi vibrations wcrc prcviousty pcrbccomc possibJc original

othcr

t!)C two periods which Convcrs~Iy, is rcmoved, constraint lie ouc on cn.c!t si()c of tho

values

of ." be cquat, Z

winch

can

on)y

iMLppen when

~=.1 h~s no on t)~c pcriod fur to one vertical pundutum crrcet

thc

introduction thc

uf :L constmint !imit~ti(jn

instanco, phme.

of a sphcrical

a. simple of a sysLem wittt two (tegrecs of cxampte of ]en~t)i H'CL'doni) wc may take a. strctehcd itsdf withstring out inortin, but cnrryiug two uqua.) nasses /?t nt distfuiccs a a.nd = 6 froin onc end (Fig. 17). Tuusion 113. rig. 17.

As

Sincc two periods If tl)e

T and

F are

not

of thc

satne case

form, nncqua!.

it fullows

t)tat

thc

of vibration loftds

aru in cvcry

thc cLa.mctcr of thc a.ttactio(~ two componcnt vibrations is vident. In the first, which wil! Itave t)Le longer titc two weights move togcthcr, se that a' and y period, rcma.in equ:d throughout the vibration. In tho second x n.nd arc in sign. Th middie point of the nmncriea!Iy cqua!, but opposcd thon rcmains at rest, and tlie two masses arc aiways string to bc found on a straight Une passing it. In the first case through = 0 x + = 0, and in thc second and + y so tliat x winch must he assmncd arc thc ncw vanahies in rdcr to rcducc T and Fsimultancousty the functions to a sum of squares.

be symmct.ricn.Hy

113.']

INTERMITTENT

VIBRATIONS.

121

For string

example, into three

if equal

th

masses

bc

so attaclied

as to divide

th

parts,

fro.n

which

we obtain

a.s th complete

solution,

where, as usual, thc constants the initial circumstances. 114. Whoi thc two Jiatural

a, 7?, j8 arc

to be dctcrmincd

by

of a systcm are nearly periods sometimes vibration of intermittent prcthe phe~omcnon equal, In order to ittustratc tins, manuer. scnts itself in a very curions with two we will now suppose, wo ma.y recur to ttic string loadcd, of th from its ends cqun.1 to one-fourth masses at distances equal were absolutcly iixcd, If thc middte point of tlie string length. th hc compictcly the two as one system, or, if thc whole be considered independent, that bc cqnal. Wc now suppose would of vibration periods to of bcing absolutely Instead nxed, tlie mid(Uc point Is a.ttachcd it is of mcrtia, so that dcstitute or other machincry, sprints, is to The reservation us to incrtia of yichling s~/t~y. capable of a third dcgrce offrocdom. avoid the introduction vibrathat thc fundamcntal it is vident From th symmctry
two sinuhu' aystons on eitlicr

side

of it would

tions

and arc thosc a?-y. rcprcscnted by a;+y system of the on account arc shghUy diffrent, bccause, Thcir periods of a dplacement thc potential of thc centre, energy yieldin~ t)iat of a disp!acemcnt is less than and v are equal, \vhcn arc ttie kinctic whcrcas x and y are opposite; whcn nergies of vibration. I)i th solution the samc for the two kinds of tlic

wc arc Now tions

theroforc let us

to regard that suppose

M'! initially

as near)y, but not f~ntc, a? aud n: vanisi). Thc

cqu:d. condi-

are

122

VIBRATING

SYSTEMS

IN

GENERAL.

[114.

wiuch

give

approximatcly

Thus

Thc prcsscd to sin-

va)uo by

of
I~u-rnouic

thc

co-ordinatc tcrm, whosc

.T? is

berc

approxitoatL'iy being function proportiona! of t!ic

cx-

an.phtudc,
harmonie

t, is a slowly

varying

thne.

TIic vibrations of thc co-ordinates are tbcrcforc so adjnstcd t]~t each iuuplitudu v~)ns)ics at tbe othcr is at its !uaximum. T)us vcry low phenomenon pitcb, hcavUy t!ic staik iuto
not

and Intermittent moment tl~at the

may

screwing in thc normal stalk

prongs ds

bc pret.Uiy shcwn fork of by a tunin~ )Lt tbc ends, i~d wui~htud fh-m!y'hdd by a massive support. W!tcn tLc fork vibrato

Le m a small

cntails
oneprong,t)ic tl~emscives

th rigidity, nanncr, or want 01 rigidity, ot' th comc into p!ay; but if' tbc of'the di.spj~ccmcnts two t)ic samc direction, t~c .s!ig)it of Hie sta!k yidding of pcriod. If t)ic furk be excitcd change by strikh~ vibrattons are and intermittent, to transfeappcar
and forward.s bctwecu th prongs. U.dc.sa,

back~-ard.s t)ie

bu vcry firm, Utc abnorm:d suj~port which vibration, a .notion of th centre of Inertia, is soon dissipated and of thon, tbe course, vibration to bccomc appcars If thc stcady iork be mcrc)y hctd in thc hand, t!ic p!ienomeuon of' mtennittenec cannot bc obtaiucd at a! 115. uscd TItc strctclicd somc string with two attaclt(jd nasses bc

howcvcr, involvcs

For example, gnerai principies. ths period of t!tc vibmt.ou which remalus wbcu onc pos~ibte mass i.s !te)d at rcst, is Intermedi~te between th two frec pericds. Any incrcase ni eithcr Joad depresses t!ic pitcb of both th natural and vibrations, If t)ie new load be situated co~vcrsciy. at a point ci th not, witb tlic places whci-c t)ie other !oa()s string cuinciding are attiiched, nor with tho uodc of one of' thc two prcviousiy frcc vibrations possible lias no nodc), (thc othcr th ef'cct is still to prolong both thc periods With alrcady to the prc.scnt. regard third nnite period, w]iicli becomes for thc first time after possible the addition of the new load, it must be as denvcd from rcgardcd

to ~lustrale

May

115.]

IMPRESSED FORCES.

123

number of which an iudennitc one of infinitely smalt magnitude, of the system. It is instructive to form part may be 8t)pposcd of a new load and its graduai to trace tlie enect of the introduction it will be but for tins purpose increase from zero to infinity, to take simpler connueiicejncnt thc thcre case where finite there is but one other. At the of inanother pcriod T~and As t!)e load increascs nnitcsimal finite, T~ bccomcs tuagnitudo T~. Let us now considur increase. and both T. and T.. continually Onc of thc becomes th load when grcat. vcry wliat happciis ail and of growing is nccessarity bcyond largo capable puriods iinite Innit. T!ie first a fixcd The otiicr must limit. approach as to a motion in which thc largi- mass vibratos nearly bcloags th second is tlie period of th vibration if tlie other were absent is one of tlie smiUler mass, taking werc fixed. place mucb as if the larber th must be aiways bc equaL Now sincc ~ and T~ can nover a~d we infer, that as tlie load becomes cot~tinually larger, gruatcr a iiuite and T~ that approachcs it is ~ tliat met-cases iudennitcly, limit. Wc 116.
frecdom

uov

pass

to tlie

consideratio!i

of forccd for a system

vibrations. of two degrecs of

Th
including

quations gnral friction arc

If
constants

thc

conncction are

between small, may in

in

th

Jt~, ~3, denominator thc

x and so that

is pcrmissiblc, from vented

co-oi-diiiate and a force varying,

gnral y is th same V had

th of a loose character, tlie tcrm (J9 ~W +1'/3~)" this bc ncglected. 'When bc as if x had bcen introduced been prewhose

But if, in consquence of N, y, and C. is independent tna~itude the force and onc of the between isoclironism of an approximate is constraincd to bc whcn x or beeome which motions possible be smaU, then tlie eitlier zero, ~+~~orC'+~ of mutual tlie coefficients in the dcnominator term coutaining innuencc and thc be retained, bcing no longer ?'e~~ue~ of a more complicatcd solution is aecordingly must unimportant; charactcr.

1~

VIBUATI~O

SYSTEMS

IV

GENERAL,

Hl~

~o.Id

~'mmetry have R~.p,

~hcw.s
fou..d

that

If we had va).c
fur

~sumed
..s no~v

the .ne

A~=0
obtah.fur.

y=~"
Th

w.

is uc lurcncd

a to as .u exiunpjo.

th.tTi~
v1 r

-suppose 1=0, case ~hcrc

<

and

vhi~ to t!.c

~t is p.-c.scnbcd, ~==~) for g~ter we .shalt coufinc .ImpHcity ourscIvcJ = U. TI~ vaiuc /3 of~ is

~<

.n. spcetivejy

''?'~

"7
~j

P~ ~~y

of t)'6

co.mcl.nt

of

Le ~e icrc

and

It appears ~hat: into tho tl.

=~

tlmt tlle be causcd .4 ci'

effect ffc'ct of tho

reaction = to

~~1" ~1 coefficients whose the of (.,

and above of (over -1 ( is ~J~~scnted by chang.-ng the ~n in

of spring cllaracter value be~g

and friction,
of

tenrylccted, Por~ 'l'jJ(3ll

a1tcmtiolls, howcver, tlce peniod of tlee ~~aotion couive now pl'occed to cousicler.
~e n.tnnd frictiun!ess

'l'liese

corrcsponding to n~u~aincd at zro);

so that

~V=o'

J J1

most

cases Wo

with sh.JI

IVhich

we are

?:?=~

acconhng!y

lrvctically of y Jeavc out

C0J1ccrno(1 'Y is nut l11ucI. ditJer.. of account ihc

117.]
variations

0F A DEPENDENT SYSTEM. REACTION

125
tcrm uot

and in thc small factor (ZF- J! of thc positive vainc ?!. Witcn its approximatc substitutc p for~) 'y~, is of no importance. nearly equal to M, the tcrm lu question As might &' is aiways ncarly, be anticipatctt positive. Its from th gnerai value which

maximum. to

of work, pnncipic occurs wlicu p = varies ~e~e~/ with

and is thcn

proportiollal

on a view of the supernclal 'y. Tins might not hve bcen cxpected th friction, a paradox for it sccms rather that, th grcatcr mattcr, tha.t 'y But it must bc remonhci'cd th !c.ss Hho)t)(! hc its resn!t. t)io and that whcn of friction, is on]y tiie co~'c~e/!< y is small maximum motion is so much incrcascd thf~t thc whoc work spent

against But

friction

is gi'catcr

tilan

if'y

were

more

considurahle.

of ~1' on thc point of most Interest is the dcncndeiicc th va.Iuc As p passes through If ~) bc less than x, ~1' is ngative. th inam~ changes WI)on J.' is ngative, ?:, ~1' vanisitos, sign. th rccovcring Hncncc ofy is to diminish a?, powcr of tbc vibration is slowcr aud wc sec that this happons whcn thc furccd vibration to Th of th vibration tliau t)iat natural y Is thus tenduncy thc slower, but to th vibration to retard x, if tho latter be ah'cady in tbc thc more rapid, ou!y vanistting accelcratc it, If it bc ah'cady to makc .B attempt with a peeuHar detcrmincd vibrate at thu rate by n is beset a hcavy in to that met with balancing difnculty, anaiogous Ou whicbof gravity above th support. body with th centre of adjustmcnt from prcision cvcr sido a shgtit may departure vibration is aJways to incrcasc of th dpendent occur th innucncc critical
case of pcrfect

isochronism.

TI~c

error. Hxatnph's will strong rsonance thc thc th most

intcrcsting

of piteh Instabihty comc across ns hercafter; of thc results of application of thc

accompanying but undoubtcdly this section

is to

of the anomahius rfraction, posscssby substances explanation of th two kinds of light sclectivo absorption, ing a, very markcd sidc of tbc on citbcr situated Immetnatuty (in a normal spcctrum) absorption
thc discovcrcrs

band*.
of

It
this

was

obsc~'vc(~

by

Christianscn that

and

Kundt,

rcmarkalde

phenomenon,

mdia solution)

of

the

kind th

in question

(for

tM abnorma.Uy ray immcdiatcly as on othcr If we suppose, it in <e/ec<e.CMs, and that above th intense is to do, that it would be natural absorption grounds
J'/u't..1~ M~y, 1872. A)so SoUm~inr, r~y. /i)t)). t. cxliii. p. 272.

in a!coho)ic example,/MC/~MC th ~~o~ absorption-band

rcfract

< 126 VIBH.ATING SYSTEMS IN GENERAL.

[LIT'.

of an agreront bctwccn. the vibrations of thc kiml of to tbc mo]eeu)es light affecte d, and somc vibration of thc proper would in(!ica.tc tb~t for light of someabsorbing age~t, oui- theory w]t~t gi-(i:ttcr poriod t!ie cH'cct inust bc thc saine as a relaxation of tho natural of the cthur, itscif clasticity rnanifustuig by a slowcr aud incrasud rfraction. Oit t))c otitor propagation sidc of tbc its rnHucucc must bc iu thc opposite absorptioM-band direction. lu ordcr for brevity, to trace 'y~ = f/, tlie jV~ law of conncction hctwecn /r) = x, so t)t:Lt ~1' and takc,

the rcsult

Whcn. serves its

the

/t' is rcverscd with it, but prsign of .<-is chan~'d, muncricai value. Whun or M, ~1' vanislies. a;=0,

Hcnce axis

thc origin is on thc of x is an asymptote. wtien

reprsentative Th maximum eclual

curvc n.nd

(Fig.
minimum

18), and
vaincs

th
of

~t' occur

x is respectively

to + ce, or a

a.n(t thcn

or 'y, th grcatcr will bc thc alteration of tuni tlio vainc of will corrcspohding uearcr a.nd nearer to n. It may be well to approach that in repeat, tlie optical application a (liminishcd is attend cd by an ~crpf<M~ maximum When the adjustment absorption. of periods is such as to faveur ~t' as much as possible, thc corrcspondijig value of a' is one hn.lf of its maximum.

Hencc, maximum

the smallcr

th

value

of

CHAPTER

VI.

TRANSVERSE

VIBRATIONS

0F

STRINOS.

118. more

MONG

vibrating

bodies

tliere

are

none

tliat

than Stretclied Strings. promineut position for musical earliest times have bcen employed thcy and in th prsent essentiel day thcy still form th parts and the vioHn. instruments as tlie pianoforte important mathematician battle-neld they on which must wcre always fouglit possess out tlie

a occupy From tlic purposes, of such To tho

a peculia.r as tho interest controversics of D'A)cm-

Bcruoulli and Lagrange, to the nature of tho bert Euler, relating solutions of partial difTerential To tlie studcnt of quations. Acoustics In consquence of th comthcy arc doubly important. of their on which parative theory, they are the ground simplicity difncult or doubtful of simple toncs, form of a Mouochord ratty available Thc
clament 'string'

questions, can bc most

such

as those

rclating

to the

nature

means

while in t!]o faccd advantageousiy or Sonomcter, afford tlie most gcncthcy of piteli. for thc comparison. is a perfectly between stretched uniform two in fixcd practico, and floxible

of Acoustics solid matter never

of Ml

pointsin though in music. devia-

fact

ideal

body,

closely approxima.ted We shaU afterwards tions from Th classes, amplitudes complete are not

realizcd actually to by most of th strings sec how to takc account ncxibility of a string and uniformity. be dividcd In

emptoyeJ of any small into two

vibrations which do

may

distinct

independcnt practically limits. exceed certain

of one

if the another, th first class t!tc

are ~o~tf~ so of the particles and motions displacements The ret~ins its straightness. that th string potential always not on the whole tension, but of a dplacement depends, energy which occur in th various on tho changes of tension parts of tho string, due to thc increased or diminished extension. In order to

128
calculate

TRANSVERSE

VIBRATIONS

0F

STRINGS.

[118.

it we must know the relation between the extcn.slon of a and tlic strmg force. Th iipproxim~tc stretching hw (giveti by Rccku) may be exprcsscd tliat th extension by s~'ing varies as thc tension, so that if aud dnote tlie uatural and t!ic stretched of a string, and 7'tlie Jengths tension,

on thc m~tcn:d and th action, dpendit ~ncti bc m~y to meaa intcrpreted tl.e tension th.tt would bc to strctcii t].e stnng necess~y to twice its natuml if t).c !cngth, law apphed to so grcat in gcnem!, cxteu.sions, it is far whicl., irom douig.
constant,

whcre

is

of U.e second kind arc that is ~YtH~~e; to say, the of th particles in planes .string movo sensibly perpendiclllar to the Ime of t), c In tliis case t)~e potential string. ener.-y of a dplacement the depends upon and th genend tension, variations of tcnsion t!.e additional accompanying ~aU stretcl.iur. duc to the dLsp]accmcnt bu Icft out of account. .nay It is he~ as.suincJ ti.at the duc to ~c inotioa .s~ching rnay 1~ nc~c.cted in co.npar~on with tl.at to ~)uch tho string is aircady il) subject its of
position e<tuilibrium. not, Once assured of th futnimcntof vibrations t).is condition, wc do iu th

119.

Th

vibrations

rcqnu-e The

to know most

investigation

of

tmnsverse

anyt)ung

further

of the huv of extensiou.

gnera! vibration of th transver.se, or latral, kind bc .y a~ve resolvcd, shal! presently prove, into two sets of comnor~ P .ent v~rat.ons, executcd in perpcndicu)ar pL~s. b.nc. it is only ill tho initial circumstances that there can be any d.st.n.tion, to the question, psent bctw~ ono plane and e~c ? sufHc~nt for to regard the motion .nostpurposes as couhned to a cntndy single plane passing th line of the tbrough Mrin~ of strings it is usual to commence ofthe di~rential partial quation of waves in the and ne~ positive thc.se in such a manner as to suit w).ose theeaseofannitestring, ire maintained at rest; ne~ther of the solutions taken by itself' boing consistent with the existence of or places of permanent rest. This aspect of tho T'cst.on and we shaU .svery emportant, fully consider it; but it of tlle treating theory with two particular solutions the transmission representing tive directions, and to combine In

119.]
aecms on a property transmission mcthod last

VIBRATIONS 0F STRINOS. 139 TRANSVERSE

dsirable so pecu)iar of waves. to found thc solution in tlie first instance as the undisturb~d to a MMt/b?'7H string Wc will procced gencral by thc more in thc with what was provcd conformity normal pcriods comand

scarcely

of assuming (in that the motion chapter)

poncnts chajactcr

into ma.y bc resolvcd of thc harmonie type, and dutorminingthcir of th system. conditions by the special carrying out tliis design th nt'st stop

Towards bo tlie thc throw

would

of thc investigation of a continuons motion liglit on a point, connection

partial

din'ercntial

cicarly,tho the passage and arbitrary what different

is subjcct. string which it is most important bctwccn finite and Innnite

equation, in But

naturally to which order to

to understand

thereto between corrcsponding we will commence functions, by following course.

and freedom, constants arbitrary a some-

120. vibration if tho out this uumbsi' centre, arrive'at

lu mass

Chapter of a string wcro

in.

it was

would concentratcd

out poiatcd not be entircty at th whole mass string of cach

th~t

thc

fundamental in charactcr,

altered

middic

idea, wu sec tbat if ttte and tho of small parts we might by sufficicntly a system, continuous

Followin~ point. werc divided into a concentrated at its

ing the lc:~t as tlie lower solution analytical its This limit is thc Lot p dnotes be the will

numbcr of parts tttc mulbip~yiu~ but capable stii of finite frecdom, ofreprcsentso far at with any dcsired accuracy, string If th vibrations arc conccrncd. component

for

givc mcthod'followcd the

can bc obtained, of divisions any numbcr to a uniform thc result string. correspoudiug by Lagrange. string, so that

Icugtb, pl tho whole mass of the mass per unit Icngth, T, thc tension. Fig. M.

Thc so that

Icngth

of tlie

string

is divulcd

into

w+1

equal

parts

(<t),

R.

130 At

TRANSVERSE

VIBRATIONS

0F

STBINGS.

fiSO.

th ?? points of division equal in~sc.s arc supposed conwhich arc tl.e reprsentatives centratcd, of' th mass of th portions of thc (~ .vlucii tl~cj string, LLsect. TI~e mass of ,.vur.y cach of' lengLh portiou is suppose.! to be concoutratcd at th term~I flua.1 pomts. On tlus we hve understand.in~

to invcstigatc procecd tho vibrions o but Icadcd at e.ch of dcv~d mcr~, aud from thc euds, witli (a) from thcmsolvc. If loadcd fuUowmg pon~t.s. <~notc thc ~tcral aud

Wc

a strin~, points ,ant a mass

of

itself

tlic initial mciud.ug fur F mid F expressions

of thc displaccmcnts ~nal poin~ wo h.vc thc

with

the

conditions Mcthod

tliat the

~nd quations

y~ish. of motion

These

Lugranges

givc

by

whcrc

now t),at the vibrat:ou Supposing ci normal type, wc assume that cos~-e).where tlien bc rcgardcd If for thc Rakc .c.uain.s constants, of brevity to bc with wc put

under

consideration &e. arc atl propor of -7~ for

is ono io a~

dctcnniu.d. a .suL.titution

tlie form

tLc

..iucs

of

..uu~cs

120.]

MASS

CONCENTRATED

IN

POINTS.

131

From

It may to sin (?~ + 1) but \ve shall sin on a hint dcrivcd from GfMC dircctly from (5) by acting results and assuming to a continuons rclating string, Titus lut a solution be particular type of vibration.

quation bc provcd th~t,

this

tho values if C= 2 cos

of the

bo found. might t!ic dterminant is equivalcnt attain our o1)]Gct with grea.tcr roots the known a for trial

'where

s is an intoger. uquations (5), we find

Substituting

tho assumed arc satisfied,

values

that

thcy

provided

of ~r in the tliat

A normal

vibration

is thus

roprcsented

by

whcre constants P,, 6, dcnote arhitrary indcpcndcnt constitution of tlie systcm. Thc w a.dmissibte found from (14) by n-scribiug to N in succession 3.W, so that values evcn~ and this arc all diHcrent. If wc tnlce to correspond of s give only tbc same periods one of thc values of Mthat~ ()oes not and of the values thc

genernl of ?! arc 1, 2,

vatucs

.s'=Mt+l, ~vttnishcs, n, possible vibration. Grcatut' over uame)y, a.ga.in. If ni + 1 bc to con'cspondiug

93

132

TRANSVERSE

vmRATJONS

0F

STRT~GS.

[1~0.

thc same as woutd bo found in thc case of un)y a=~ (~ + l),is is obvions. Jn tho kmd Thc interprtation n. single load (~ == 1). rona.ins nt rcst, so of vibration considcred partictc every n-lterruite wo'u oncs as titough the intermediate that thcy rca.Hy movo of struigs of Icngth at to tlie centres a.tta.ched 2c< fMtcncd the ends. Th possible most solution is funnd general solutions of norinul particular hy putting type togcthor a,Ii tlie

and, by ascribing identificd with cumstanccs. Let a; dnote

suitabic thc

values

to tbc

vibration

resulting

can bn ar1.)Itrary cn~~iants, from arbitrary Initit cir-

tbc

distance

of the then may

partic~o

f from

thc

cn(~ of the fur~. unda

so that ()'l)ct=x;; striug, from (1) and (2), our solution

hy substituting he written,

we hve to p:ms to the case of n. continuous strittg, Th fn'st Qqnn.t.iou rctains its form,)') 1 put ?~ induite. oniyt') at any point a*. ThL: tiMiiting furm oi RpcciHes th disptacumcitt ttie second is simply In order

The that whole

of tlie compnncnt toncs periods of the series, of tiie gravcst fuund motion is in a.U cases periodic;

arc

thus

alicpiot

parts

uf

by th

puttin~ pcriod

N=1. is 2~t/

Thc

and

This statement, a shorter pcriod Jower compoueuts

must not bc undcrsiuod as cxcludi!)~ however, for in particular cases of tl~ any uumber is that, ti)u Ail that is asscrtcd may bc n.bsciit.

120.J above-mentioncd

MASS

CONUENTRATED of timc

IN

POINTS. aboutacont-

13~

interval

is 6'M~tCte~ tobring

W';t!<f")'.))'th discussion p!(.'tcr(:cu!<)~?'\ prusentanyfurthpr formuJn, oftttc to observe the (1!)), but it is ititerusting import:t.~t to a limit iti (17), as ?~ is madc Hucccssively and approach grca.ter For tins pm'poso it will bu suHiciunt to takc thc gr~vest ~rcuttjr. f).nd according)y to trace tuuc for w))ich s=l, the variation of ?!2(w+1) n/' ,1~' TT 2(M;+1) Thc tion fuilowing aud variable
?~ ~) -T sni ~(~t.+l)

arc

a sries

of simnitancoua

values

of

tho

func-

I .9003

2 .9549

3 .9745

4 -983C

19

39 -9997

.995U

.9990

values of m th limit is for very inodcrate of (!novca.ble) ?~ is tlie rnnuber Sinco loads, closely approachcd. to thc probbiti in Chapthe case ?;= 1 corresponds uive.sti~ated wu must rememher thn.t we th results ter but in comparing 111., the w/~e m~ss of tlie string to bc concentrated at thc tliere supposed It will bc sccu that centre. grca.t; wbere In th thc
prcscnttCasc

thc

h):).d n.t th cuntre couecutrated

reina.indcr

it is witliout th

bcing cf~ct.

supposed

is oniy haf as at tlie ends,

Froni initial

fn.ct that of the string

our solution

is general, hy

it follows

that

any

form

c:UL bu niprcscntcd

fur thc stringataU And,su)co auy furm pos.-ilbio we infur thut any iini.t.e singlu valued rc-g~rdml as initud, ~nd a;=~, c:ni be exp~nded uf te, wltich v:).)M.s)ies at ~=0 those Ihnits in n. scries of sincs
t)icorum.

mn,ybc functioti withiu

of

a.nd

its

mu)tiplcs,whtc)i sitcw how the

is a. Citsc of Fouriur's more gcncml We furm

Vc sliall

prcscutly

cnn bc duducud. now dtermine the constants for a. continuous to solve afterwards the to

121. string

might

a~ lu !)3, but it is instructive by Intcgrntio!i first in tho gcnct':d c:tso (~ finitc), aud probicm are TIic iuitial conditions procecd to the limit.

134

TRANSVERSE

VIBRATIONS

0F

STRINGS.

[121.

where, arc tlic To sin s~,

Lrevity, ~~=~cosf,, iuitial displaccments dotcrnano thc second a.ny constant

for

and ~-(r~), of th w p:u'tictcs. muttiply &c., aud of a!l the Henco

'(2a)

i~(mM)

t!tc <t.dd tlie

first

quation

by Thcn, J,,

by sin 2s thc of coefficients =

results. cxcept

byTrigojiomctry, vanisli, wliile tliat

constants,

(~~ + 1)

of 7?, (cqu~l !)orc to write down the values W!tcn becomes on the initial vcincitics. to jf~, sin e,) ibs deponding tho sign of sutumation ranges by in<ismaU, )'~ under I)i~nite]y 'Wc ncc'd not stay nitcsinial so tliat stcps writing from zero to At tlie same time '??t + i = ai t

?'M = x, f = (1,,v, we Iiavc

u!tima.tc!y

cxpressing 122.

d, lu tcrms

of thc

iultial

displaccmcnts. the partial difFercn-

Wc wi)t now invcstigatc

indcpcndently

HcxiLfc ti:i] quation govcrnin~thctt'ansvo'.sumotiottofa.po'fcctiy ofthe tension on th suppositions jnagnitudc (t) thatthe strin~, of H)c inclination constant, (2) t!)at thc square mny bo cunsiucrcd of any part As befure, constant and of thc string dcnotus thc Let
to Its itntial

diruction

may

lincar

dcnsity

tension.

rcctat~nlar

at any point, bc takcn co-ordinatcs tite

bc ticgicctcd. and y'~ is the pandie!,

su t))at x: givcs to thu stril~, pcrpcm~cuhtr and .c, y, z thc disptacud p<jsiti<'n uf any partictu (/.c :u-o th tensions on thu clment forces acting und any impresscd forces .)~ ~p< ToJhuutor'H J)t<. C;c., p. 267.

cquilibnum Thc at tinic t. cuds, Pnn-

at its two

ByD'AIcmbcrt's

122.]
cipin against ponents thse form

DIFFEBENTIAL

EQUATIONS.

135

an

equilibrating p CLu p MC

system At

with tlie

th

ractions x thu com-

acclration, of tension arc

point

If thc
on

squares clment

of

(a;

c~x arising

Le ncgiccted out of tlie

so that tension arc

ttic

forces

acting

thc

IIonco

for tho

equations

of motion,

from

winch

it appo~rs indupundont

that

attogethcr Tho spoudmg be written

dpendent of onc another.

th

variables

y and

z arc

student cquations

should

tlicso compare ofHuitc diM'crcncGs

quations in 120.

with Thc

the latter

corrcmay

which

nmy

:).!su bc provcd

dircctiy.

13G

TRANSVERSE

VIBRATIONS

CF

STRINGS.

[123.

Th second,

nrst

is obvicuii

from ~o notice work tension

thc

deHnition thc to

of potcnti~

2~

it is sufHcipnt is the configuration stretching against tion of equihbrium, th

tha.t

prove in cncrgy

To

the a.!iy

requiro~ T,.

tlic producc from tlie Ruckoning

nceessary conHgura.-

wc ha.ve

and,

so far tta tlic

third

power

of

123. tho

In most

of thc

dctiatty p Is constant, motion may bu supposed write thon convenicBtIy

applications there arc to take

that no

we sha.H have

imprcssed in onc pheu

to mako, forces, and the We may plane.

n.nd tlie

difrerentia.!

cqua.tion

is expressed

by

If

we

now

assume

t]iat

varies

as

cos ?)M~

our

equation

bccomcs

of which

the most

gcnera.1

solution

is

This, thc
assume

is uot thc most goieral howcvcr, lu ordcr to obta,in in question. period

ha-rmonic the lattcr,

motion ws

of must

\vhcro
substitution

;?/
in

M'c fuuctiotis (2)

of

a*, not

tions

uf th fut'm

it appca.rs thn.t (3), so tlia.t Hnally

ucccssarUy arc y~ a.ud

thc subjuct

samc.

On

to cqua.-

:ut expression tiuuous tcu~tL

For any conconstants. four a.rbitra.ry conta-himg the differwithout of string sa.tisfyiug iutcrruptiou

123.]
ential cquaticn, th condition, monie. But this
th~t th

PIXED

EXTR.EMITIES.

137

is tlie
motion

most
at

whenever

thc

string

solutioR under possible, gcnera.1 harevery point shaH be simple forms part of a. system vibrating from former wo know chaptcrs in thc same which phase, r\

n.nd withoub frccty aru t,)):it :).I1 parts that t'cfjuircs then thc most

dissipation, simuit~neousty

and

goncra.1

vibration

of simple

harmonie

type

is

as wcH as t)ic most nnpo-tant problem of th free with our prsent subjccb is the investigation connected fast at both its ends. held vH))-~tions of a fnntc sti-iug of Icngth a,rc conditions If we takc thc origiti uf a.- at ono und, tlie tcrmirnd for ~11 values of t. vanishes Mid wheM a!=~ when that a:=0, T)ie nrst i-cquit-es tha.t in (G) of 123 12-t. Thc most simple

and

th

second

that that thc only

We IcM-n wlicre s I.s ~n intcger. ~==.S7r, are such as mnkc vibrations hM-mouic possible or that

iUtdthuu

may bc, it htu-mouie a.nd vibrations, as a suni of simi'ic be rcprusoitcd CMt solution for a string, th:tt th luost gcncra.1 coucludc wc thurcfurc Now wc know M./)Worz th:tt whn.tcvcr th motion

138

TRANSVERSE

VIBRATIONS

0F

STRINGS.

[124.

ail is under so that, as has bccn aiready the whoc motion statcd, conTh sound cmitted circumstanccs in the t:mc pcriodic r~. of stitutes in gnerai to our dennition a musical 7:0~, acconUng of its gravest is nxed by the period that term, whose pitch cases that the in special It may happen, however, component. is not is absent, and yct that the whoc motion vibration gravest periodic ~/+~/ finite. in any
shorter

tune. for

This

condition

examp)c, ~l./+7?./ a note; lu such en.ses the sound could hrn'dty be called but it usuiJIy in practicu tone is w]tcn tho gravcst that, h~ppcns .some othcr takcs its p)acc in the cbaractcr of fundamcntal, absent, and the sound still constitutes a note in the ordiltary sensc, though, the odd whote of course, of c!cvaLcd beginning compollcnts motion is thcn missing. in the tit-nc ~Tp and if the periodic bc prsent, th sound prsents nnusual. nothing pitch. with
A simple case is wheu ail

vatjish,

while,

of things occurs, if and ~t~+~ are

thc

first

are

Tho second

component

T]~c pitch of thc note yicidcd (C), aud thc character by a string of the fundainenta! werc first invcstigatcd on meclianical vibration, in 171-5 but it is to Daniel Bernouni principics by Brook Taylor wc owe the gnrt solution containcd in (5). He (175.')) tbat of particnlar obtained soluit, as wc bave donc, by the syutbusis in accordancc with his Principtc of the Cotions, pcrnussibic existence result opinion
th scrics

of Sniat! so arrived of Eu!er,

of sincs

Motions. at was and

in

In

bis

time

tbe

opcn to question; aiso, strangdy cnough,

(;")) was not capabte

gcncrality in tact, it
of

of the was tlie an side,

ofL:t,grange',that rcprescnting the other

arbitrary drawn was

function; from the

and iunnitc

Bcrnouln's nuinber

on of thc

disposabic

constants,

certaiu!y Most of the

inadquate~ ]aws embodicd in Taylor's (C) had (1G3L!) by Mersennc. formula been Thcy

discovcred may

experinientaHy bc stated tbus

longbefore

SoQRiGmfU)D'ajPr<<f~<; Dr YounK, iti Lia momou'

D~/c'rctXtn! of 1800,

O/t'tc/tftN~c~, HC-iHH to liave

78. understood

this

matter

quito as

<orrcct)y.
nvory

Ho s~'H, "At tlio samo timo, ns M. DernfXtUi tma ]HHtIy obsorvod, Rinoo
bo iu~uitoty may tho ot'dinfttos of demonstrntod in tho samo approxinxited, au iniinitc thttt timo aU tbat aud mnuber tbcs a by of ctt.sidunnt; tmcix'id.s its of ortiinutofj tun~nirevert trochoida! to (liFfcrcnt woulJ into a

ligure of <'u)nposoJ

it may bo tUticH, tLicir initia) Htato, curvc wouhi purforn)

cunstitnont Rimiln.r this is choni in ttoinc

ou'ves bcnt

a sinn)o

'vibration

retipecta

a couvomoat

oud eumyoudious mothod of consideriug

tho problom."

124.]

MERSENNE'SLAWS.

139

(1) For the length. This pears (2) inverseiy

a, givcn

string

and

a givcn

tension,

the time

varies

as

is the

fundamcntal bccn understood tho

square

to hve Whcn as
thc

of th principle by th anciects*. of the

monocbord,

and

ap-

length
rout

et' tho

string is given, tension.

the

time

varies

of th same length and tension vibrato (3) Strings w~~ich arc proportiona) to t)tc Stmare roots of thc lincar
Thcse dimensions, important rcsults may

in timcs, dcnsity. mcthod p, and of 2' rcare

aH bc obtained tha.t T dpends time and

if if thc by

it be assumud units [Z], ~=M, [2'J, of

by the on]y on mass be

Fur, spectivcly givcn and hy

length, th [~j,

denoted symbols

dimensions ~=[~L~], of

of thse

p=[~Z-'], onty combination Th

thus

prcscnting is Uic numeriea.1 125.

ments.

(see 52) the a time is T, factor.

thcm

oniy thing

of recapable left uudetermined

Merscnnc's In playin~

laws th

are violiu

cxcmphfied din'ercnt

a!l stringed notes are ubtaincd la

in

instrufrom

its cnicient string hy shortening Icngth. vioun or the pifmuforte, an adjustment uf pitch a constant t!ic tension but !engt.h by varylng mcmbercd tliat /) Is not quite invariable. To secure rcquisitc thickness, latitude. Vnicncc,
an incrcase

thc

same

tho tuning is cectcd witli it must ho re-

a prescrihcd
onc rctation

pitch
only

with
bc

a string'

ofgivcn
bctwccn

that and Th and

of

satisficd

materiaL the Icngth,

it is tiie

th its to

but tension; is often length curtaiimcut bccausc, thc

in practice thcrc is usuaUyno grcat limited of conby consi<turations cannot be compensatcd idways by if thc tension he not loss of increascd HcxihiHty, is cH'cctcd towards strings wirc,

thickness,

proportionaDy whUcif'thc

tension

the pitch. lowering ofUic and violin pianofortc whose cU'ect is to Impart ncxibility.
Aristono which "hncw t.tmt a tbovibmt.iousoccupitid on tho JopeudoJ pmport.muH

thcro is a section, bc so incrcascd, nothing T!ic dirricuity is avoidcd by thc addition Inc'rtia' wiLhout

in t!tc )owcr of a coil of fine too much

impairing

of dnohi )t ftonud of pipo or (t ohnrd Jen~th pt'oduco'l a JuuH [md timt tho propcrtics of coiteords timo; of tho thnes of tho occnpiod by tl)0 vibrations

soparftto sounds.Youu's

Lcetu~M o)t Ntt<xnft~/tt<uM~y,

Vol. i. p. ~01.

140 For

TRANSVERSE

VJBRATJONS

0F

STRINGS.

[125.

into the laws of quantitative investigations the strings, is emphjycd. aononcter of a, weight over Hy mcans a lianging or a mctaHic across two bridf-cs wire, is stretcijed puUey, a catgut, on a rsonance tnounted A moveable case. bridge, whose position is cstimated to thc by a sca!c running parahel \vire, thc
i-ivos

means dcsit'cd in thc

violiu.

of

shortcning

extunt. harp, or

of tlie wire to any portiott TIie vibrations may bc cxcitcd as by p!uckin" witli a. bow (well suppiicd with as in tit rosin),

tite

cfHcicnt

If the nnes, thc

one-third,

be placed bridge ha!f-waybGtwecn note is raiscd an octave; whcn thc string th note obtained is tt)C twclfth.

moveable

the is reduced

Dxcd to

of the law of lengths, Mcrscnnc determined By means for thc nrst time thc frequencics of knowu nmsicul notes. He adjusted the Icngtil of a string until its note was one of assurcd in th positiuu musical it under t!)e same tension scale, and then prolonged until th vibrations were slow enough to bc couuted. it is convenient to hve exprimental purposes two, or mounted side hy sidc, and to vary in turn more, strings thehto winch they are Jcngt!i3, their masses, and tlie tensions subjucted. Thus In order that two strings of equa! length may yle!d t))c intcrva! of t)te octave, their tensinns mnst be In thc ratio of 1 4. if th masses be tlie samc; if thc tensions be the same, th or, masses must bc in thc reciprocal ratio. is very uscfut for thc nmnerleal dtermination the tension, tlie string is tuned ofpitch. By varyiug to unison with a fork, or other standard of known and thcu frcf~ucncy, by of th moveable thu of the strin"' adjustment bridge, is Icngttt whieh vibrtes in unison determined, with for any note proposed mcasuremcnt. Tito of Icngths ]aw tticn th mcans of givcs cn'eeting Anotiicr absente t]ic dL-sired
comparison of

For

Thc

sonomctcr

frequcncies. tho dtermination is tlie samc as on of that

cxptainud tiiu absulute ?'a~o and

to application by Scheib)er is Important. Th priucipiu pitch in CIiapter ni., and thc mcthod pitcii th of two notes from (/~(;7-e7ice
in

dpends

of their
unison

a knowlcdge Th frequencies.
with

deduchifr of both t)ie of t)ie Icngths of th

souometerstrmgwhen

it four bats lungths

p'u- sucond, are cafuUy is th iuversc ratio of th frcqueucies,

fot'k.andwhengivin~with mcasured. Th ratio aud thc

difierence

125.]

NORMAL

MODES.

141

ofthe th

isfour. frcquoncies fork can bc ea.leula.ted. pit.ch of a string from tlie mcchamcal are necessary

From

tbcsc

data

thc

absolute

pitclt

of

Thc mu]a cautions

may be calcnlatcd of tlie eicmcuts

a!so by system, Thc tonsio))

Taylor's but grcat

forpr-

is producc<) a.ccuracy. with tl)o samc unitasp) m:t.y bo (cxprcsscd by a.welght,whosemass = 32'2, if th units oficngth a-nd called P; so that whGi-e y, = In order to securc that tho who)e timc bc the foot aud th second. no bridge must bc intf.;]-acts on t!tc vibrating tension segment, tlie string a condition by suspending only to bc Siltisfied poscd, a portion of th string thc weight is !Lttachcd, After vcrtiea.lly. is and tlic length it nrmiy at two poitits, is isola.ted by dumping refers to tbc strctchcd mea~urcd. The mass of the unit of longth

to secure

thc and may bo found in<Urcct!y by obscrviug string, as of the same order of magnitude due to a tension elongation to what -\vou!d be produced and by T, a-ccording calculating of th string a. known length in its Hooke's law, and byweighing carc hve bccn sccurcd After the clamps stato. grca.t normal statc of the is rcqnircd to inftucnce resttits. Whcn 13 at a string any vibrtes in its gravcst to normal '7T.7; stn But incrc~smg no intcrmcdiato it is othcrThus, if the mode, tlie avoid fluctuations In th tension. of' tonpcra.turc, tliis way Sccbcck which obtaiued wotdd very

scriousty ~cenratc 126. excursion

moment

proporttouf).i

nurnerically point ofthu wiso in tlie vibration

from eithor end towards rcina.tns pcrmaucntly string case of thc mode ingher normn.l by bc of thc cxprossed

the centre; nt rest. componcnts.

t] le

excursion

S7rX is proportional into and

to

sin.

1.' l l whichvanishesat~1 parts. Thse

points, motion fast tion of

thc string dividing arc caticd nodes, in any way harmonies'

s cquat

without

Le rna-y cvidcntiy the vibration. disturbinp;

points of no or Iield touci~cd T)tC produc-

at thc points of thc string toucliing by iightty Ail rGsource of thc violinist. is a well-known division aliquot hve not a node at thc which modes are excludcd component so that, as regards pitch, tlic cuuct is the same as point touched; thcrc. fastened if tho string werc securely

142
127. dpend exprcsscd

TRANSVERSE

VIBRATIONS

0F

STRINGS.

[127.

Tho constants, which occnr on thc epcci.i! cii'cn)nnt:mces in tcrms ofthc initial

h) the gnerai value ofv, 124, of t~ ~))~ :~t;y b' \)i.t.t!ot), of~ a.ud

valuus

Putting

t = 0, we fmd

Multiplying

hy

sin

and

intcgrating

from

0 to

wo obtain

Thcsc dpends

rcsults

Stokes' cx(jmr!i(y on thc mitia-! vctocities,

ifiw, 95, is

for tha.t part

of~,which

and

from

tl)is thc by thc

infcrref!,
stitutmg~for~.

dcpcnding wit!i diH'ci-eutiating conditiou of the

part

on initial respect

displaccmcnts to the tirne, some us

may bc and sub-

Witcn

thcsc known, thoronghiy inotiou ibr ait subsquent timc. at rest, aud so displaced initiany = 0, and triangle. Then

at string formu!:L! allow For

one to

moment c{dcula.te

la the

]ct tho strLng bo exemple, that it forms two sidcs of a

on intgration. Wc nodc sec that vanishes, if' sin~ =0, situatcd bo aitbrdcd that at j~. if thcre A more be a commode

of thc componcnt iu question view of tlie subjcct prefensive will of solution to bc given prcsently.

by another

128.]

POTENTIAL

AND

KINETIC

ENERGY.

143

128. tiio note normal thcm

In

tlie

expression of

fer

thc

coemdcnts iv. M]d v.

of sin We wi)I

arc de-

co-orfimatcs thcrcfofG

by of tho System at any instant to the quations according

Chn-piers so thfit the arc dcfined

an<] motion conjuration of d~ tmd by the values

to form thc expressions procced for to dcducc thu no)'ma.I eqnatious of vibration. For thc kinctic ciiergy,

Wc

aud

1~ aud

Lhcneo

of cvory of product pair of normal co-ordiua.tcs. proporty

th

tcrms Hence

vanishing 0

by

the

gcnend b

do not presuppose expressions any paj-ticular either jnotion, or othcrwisc but we may apply natural, thcm to calculate tho wi)o!e energy of as follows :If string vibr~ting j)~' nattirally, bc tlie whoc mass of tlic string (pl), and its cquiv~cnt (n~) bu substituted for we find for the smu of th cuergies,

Thcse

144 If the nK'rc'Iy plane. Lagra,nge's

TRANSVERSH motion th bc not

VIURATIONS connncd of thc to

0F the

STUINCS.

[128.

to add

cno'gy

vibrations

wc havn p):u)c of in tbc pcrpcndicuiai'

metbod

givos

immcdia.tcly

thc quation

of motion

which initial

hn~i hecn ah'cn.dy values of 6 and

considcrctl tlie guncral

in

GC. solution

If is

<~) a.ud

bc tho

tit work do])c <I~ is such that <I~ 5~ ]'cprcsc!)ts By dc~nitio!i forces on thc dispI.Lcemcnt if thu Hcncc, by thc imprcssud 8~. fut'cc acting of tlie strmg p at tirnc ou an cioneut bc p 1~

is a. tincar qnaatity, quations as \ve scefrom a force of th ordinary kind. <I~ is thct'cfore In theso 129. unprcsscd hourilood a whoc, In tlie that a.pplica.tious will be supposcd to point .K=6, and wc act may

(1);

and

force of one so that

Ii~vc to make, the only in the immediate neighbe rcckoned as usually

of thc point of application tbc mode (~), <I~,=<), and wc Icarn on tho componcnt. without influence is of grcat it shcws, for importance at l'est whether auy the of

If th

force that

eoincidc the force

witli is Tins

a. node aRogether

of

in question.

that exarupic, of cquilibrium, in its position no force applied in th form of plucking, or bowitig, striking', th even its point normal componcnts'. be 1. If aftcr datuped, tlie of application

principle if a string bc at its centre, can generate opration touehiug of it

force,

as by

1 The obaBrvation that a. harmonie poluta ia plucked, iti duo to Youug.

!s uot gencratcd,

whti ono of its uodnl

129.]
with th ponents finger, which have

YOUNG'S

TIIEOREM.

145

aH motion not

must a. node and

tl~oso

forthwith at

cease

for those q~stion are absent ht

coma.re from

of a gencraHy, by damping any point vibrations which have sounding string, wc stop :dl the composent not, aud Jeave cntirely unaifueted those which ha.ve a nodu at tlie touched. point The let case of a string aside at one point and afterwards puticd rest may Le regard cd as includcd in th preceding Th complete solution thus. Let may be obtained commence at th time <=0; from which moment of at time t is

stopped Lyttie thc bcginumg'.

dumping, More

t! point, wbich hve,

go from statements. the motion = 0. Th value

where affected and (~).

(<~), with

(~)~ dnote th suiBx N. by

tlie Now

iuitial in tlie

values problem

of

the

qua-ntities (~ = 0~

in ha.nd

is determined

if y dnote point b.

th Hclice

force

with t

which

the

string

is he!d

aside

at the

at time

..(5), where The = s~ra symmetry of the of 107. expression (5) in x and b is an examplo

of th principle The

of determining the subsequent motion of a string problem set into vibration at thc point b, may be by an impulse acting treated in a similar manner. (6) of 128 over tlie Integrathig duration of thc impulse, we find ultimately, with th same notation as bcforc,

A liko rosutt ensuos whon th point which ia dampod i.-iat tho samo distance rom ono eud of tho string as the poiut of excitation ia from tho othor on!. R. 10

146

TRANSVERSE Le denoted t

VIBRATIONS

OP STRINGS.

[139.

if~y~ (2)atHme<

by 3~.

At

the samc

time

(~).

= 0, so that

hy

The than The

sries

vibrations is less convergent ofcomponcnt for a, struck for a plucked M the prcceding shcw. string, expressions reason is that in thc lattcr case t!ic initial value of y is and makes only IV cliscontinuous, spring. Sce wlu!e in tlie former it is

continuous, y itself The that

a sudden

32, 101.

of~ set in motion problem string hy an impulse may also be solved by tho gnera! formuJ Tlie force (7) and (8) of 128. <mds tLc string at rest at < = 0, and acts for an short infinitely time from ~=0 to ~=T. Thus (~.). and (~). va~ulsh,and (7) to of 128 reduces

tho disturbing force to be consupposcd centrated a.t a. single poi)it. If it be distributcd over a distance on citlier side of we l)avc only to iutcgratc th expressions (C) aud (~) with for cxample, in (7) in respect to substituting, r tTplace of .1, sin -y,

Hithcrto

we

hve

Tho tbe series

principal

effect

of th distribution

of t])C force

is to render

for y more convergent.

130.] 130. that Th problem

PIANOFORTE which will

STRING. next engage of t!ic our

147

of th~ p!ancfnrtc wit'. thc blow of a hammcr, wiiich after th impact rehounds.
assuming, as in th iast section,

attention is Thc causu vibration la hcro is projeeted tlie string, and against But we should not bc justified in
that the mutual action occupies

so short tlie very

a time

that

its duration

standards

small, periods of tlie string. of a pianofoi-to arc express

purposc of

of ordinary but hre thc

Mea.surcd may be ncg)ccted. by life tlie dnration ofthe contact is Indecd

comparison is with

propcr Now tlie with

tlie

natural th for wires tho

hammcrs svcral

used layers

to strike of eloth

covcrcd

them more yielding, with the effect making of the contact. The rigorous treatment prolonging of th problem would bc difficult, and th solution, when obtained, too probably to be of use; but by introducing a certain complicated simplification Helmholtz has obtained a solution all the representing essential features of the case. He remarks that since th actual of the string must bc slight in comparison with that of yielding the covering of tlie hammer, tlie law of tlie force called into play the contact must be ncarly thc samc as if th string during wero nxcd, in which case th force would vary very noarly as absolutely a circulai' function. We sliall tlicrcforc suppose that at the time t = 0, whcn there are neither velocities nor displacements, a force and continues .Fsin~ betiins to act on th string at a:=~ through half a period of the circuiar tliat is, nntil <="7r-jp, function, after which th string is once more frce. Th magnitude of ~) will on thc mass and clasticity but not to any dpend of the hammcr, with which it strikes tlie string. grcat extcnt on th vulocity T)io i-cquired in thc gnerai solution formula is at once 128 obtamcd by substituting its value given by for

(7) of

148

TRANSVERSE

VIBRATIONS

0F

STRINGS.

[130.

and their

th

final

sohttion

for

becomes,

if we

snbstitute

for M and

Ya.}nus,

We sec tha.t, a)! point

componcnts

vanish

w]uch

ha.vc' n. nodc not

of excitctnent, but !:(.w of force. particu): in thc infortnation

this cnnctusion Th Intcrest

does uf th

H)!).t may be ctieitcd frnm enco of th rcsulting un thc duration vibrations of contact. If wo dnote the nitio of tliis q~antity to tlie fundamcnta! of period tlie stnng so tha.t = Tra 2~ th expression for th ampliby tude of the cumponcnt s is

dpend solution lies prsent it a.s to the depcnd-

at the on a.ny

Whcn ' grcat,

in

nnitc,

wit)i N' Some tUbwancc rnust at.so coivo-ges bc ))i!idG for tho (hnte breadt)~ of thc btunnicr, thc cHect of which will a!so bc to faveur th convergence of thc sries. Thc laws of tl)c vibration

?' thc sries

those components disa-ppcar, t.~c duratinn of contact; and

wbosc wltbn

perlons .s is vcry

of strings at least Tnay be veriHcd, in their main featm'cs, by opticiil mcthods of observationcither with thc vibration-tnicroscope, or by n. trn.cing point rccoi-ding tlie characteroftttc vibration on a revolvmg drum. This character on two tbings,thc mode of dpends a,nd the point cxcitement, whose motion is se)eetcd for observation. Titosc components do not appear winch bti.'ve nodes either at the point of cxcit.cmcnt, or at tbc point of observation. Th former are not gcno-atcd, and t!)C latter do ])ot mfmifust. thcmselvcs. Thus t!ic himpicst motion is obtaincd thc string at the centre, and obscrving by ptucking une of tbc points of trisection, or vice w?'M. In this case t!te first
vibration

harmonie
is

wbich thc nf'Lh

contamintes

th

purity

of

thc

principal

sunictcnt shall

t.o prudnco t)ic results compare

wbose cornponcnt, nmch disturhancc. of tiic

is usuaUy inintcnsity In a future wc chaptcr with aurai dynamica.! tlicory

130.

FRICTION

PROPORTIONAL

TO

VELOCITY.

14S

but rathcr with tlie view observation, thu [aws of iimm~, LiuUi uf confirming' 131. solution
jnethod,

of disc~vcring and Lhu theury Itseli'.

tcsting

of a. penodic force is Included in t!) general to foUow a somcwha-t dirEcrent of 12!S, but we prter lu to m:Lkc fui cxtcusion ordcr in anc'thcr dircetion. We takoi no account forces, but wc will ofdissipativc ofca.ch lment of th string is resistcd its velocity. TIte partial dinercntia!

Th

case

have

hithcrto

now suppose that th motion to by a force proportional becomes quation

by means simptor to

of whieh avail

suhjcct may bc of th rcsults oursetves case the

the

trea.tcd. uf th

But last

it is still chiipter, reis of

that in tho ma.rkmg prsent the s:unc form as T. In fact

fnctiun-function

wherc y fmd motion

<

< are

are

thc

normal to

reduccd

co-ordinates, sums of sq)t:u'o.'j.

by means of which Tho equntiot)s of

are thei'cfore

simpfy ~+~.+~.=~(3),

of thc frcc<)o)n. ter tive

ni.,

samc

form

as obta.ins

for

subside

It is only ncccss:u'y to that sincu K is indupcndunt in suc!) a manuer that t!tc

with but Systems add to what was of

amplit.udcs

one said

of dcgrcc Iti Citapvibnt.tions thcir rcla-

thc

jmtural
manitidu

values. If a periodic force .Fcos~ act at a single point, wc liave

thc natural vibrations thcrc bc any among one ncarly isochronous with cos~< tLcn a large vibmtio)i of th:ht ty})o will bc forcer, unless Indecd thc point of Gxcitcment s)iould to happcn

If

150

TRANSVERSE

VIBRATIONS

0F

STRINGS.

[131.

fall
v:brn'-)

nea.r a node.
~n .i'j<<)~

In the thc is

case of exact
n<

coincidence,
fo app]id

thc
!tt

componcnt
a. nu'-iu ca.u

VMmshc.s;

gcneratc may be friction,

it, under rcmarkc(), /<:= 0, and

prsent very

law

special

of friction, in character.

whieh If

howcvcr, tliere be

it no

132. normal course and

preceding co-ordinatcs in to free vibra.tions

The

they whcncvcr ultimately examples. In tlie

of the use of example a probicm of forced It is of vibrations. that titcy are more cspcciaHy applicable, bo uscd witli may gcncrally advantagc throughout, the system after th of various forces is operation left to itself. Of this application we have already had

solution

is

an

of vibrations due to periodic forces, one advantagc of the use of normal co-ordinates is the facility of comparison with th efir:<(?~ it will Le remcmbercd is the theory ~<?o?-~ which of th motion on the supposition that th inertia of the system If the value may bc left out of account. of thc normal co-oron thc cquilibrium bc A, cos~, dinate then thc actual theory value-wiH bc given by the quation

case

so that, whcn thc result of th equilibrium is known and theory can rcfidiiy bc cxprcssed in terms of thc normal the co-ordinatcs, true solution with thc effects of inertia included cn.u ~t once bc written down. In the act at tlie prsent instance, b of thc with if a force string, whieh the will .Fcos~ tho result string bo of vcry of the would long period

theory, consist

point in aecordacc of two straight

equilibrium a.t any moment

portions,

132.]

COMPARISON

WITH

EQUILIBRIUM

THEORY.

151

from writing

which

the

actual in place

result of tins

for all values

of p i~ derivcd

bysimply

Thc

value

in cxprcsscd funte expression forni tlic quation

of finito

in

and and

similar th

cases

tcrms,

is usua.IIy of the normal functions of motion

tlie difHculty no greater than that of findin"th wlien tlie systein is frec. Thus in

ma.y Itowcver of 0'btaltnng

be

and

a subsequent ditions. In thc only

dtermination probicm

havo

to supplemont function compicmentary function, is of the

conboundary of forced vibrations ??t is given, and we solution of (3) with' th any particular two arbitrary constants. This co~taining

of ?~ to suit

thc

and th ratio of tho constants apart from tlie value of same form as thc normal and a.11 that remains functions; to be enected is the dtermination of the two constants in accordanco with th prcscribcd conditions whicli tlie complete bounda-ry solution must satisfy. Similar considrations in the case apply of any continuous system. be applied at a single point, there are two distinct to be considcred; the first, whcn at th problems force acts; tlie second, when It is th point = &, a given periodic actual of tho point that motion is obligatory. But it will bc convenient to treat theni together. Thc usual differentia.1 equation 133. If a periodic force

is satisfied b, but

over both

is viola.tcd

parts into which thc string in crossing from one to th othcr.

thc

is (UvIJcJ

at

152
In must introduce of junction. (1) (2) imprcsscd Thus, order

TRANSVERSE
to allow assume Thse there arc for

VfDBATIONS

0F

STRINGS.
arbitrary for

[133.

thcrofore tlie

a change in thc distinct expressions whidi must

wo constants, and a,ftcrwa.rd8 at thc point

two conditions

bc satisfied

Tha.t That

is no discontinuons of the

change acting

in thc

value

of the

thc rsultant

tensions

at b balances

force. IfFcos~ bo tho force, th second condition gives

where

dnotes the altration in the value \.a~/ (") in crossing the point x = in th positive direction. We the sha!
t!ie

of

f~'

Incurrcd

however, exponential
symhoHcal

Hnd it e"
solution

part, tion timt~

complex when

advantagcous a.nd 6tia!Iy

is completed.

to replace tho disc~rd

cos?~

by

imagiuary On the assump-

varies

as e"

thc differential

quation

becomes

of (3) consists of two tcrms, proand cosa;; Lut thc comlition portionn.irespcctively to be sa,tishcd a.t ~= 0, shcws tliat thc second ducs not occur here. Hence if y e' be tlic value of at x = b, genera.1

The

most

solution to 8ui\a;,

is the to x;=

solution In

to the first part of tlie n.pplying string from a;=0 likc manner it is vident that for ttte second part wc

sttaJtlia.vo

If y bc given, of thc problem, further to kuow

constitute the symbolica.1 solution quations but if it be thc force that bc given, we require thc rcla.tion betwecn it and

thse

133.J

PERIODIC

FORCE

AT

ONE

POINT.

153

D~erentlution analogous

of (5) aud

(G) and

substitution

in thc

cquation

to (2) givcs

Thus

Thcso proved duc to had thc In

quations in the last force acted force

excmplify

th

gnral

law that have tlie

of been

th

chaptor; at & is th at x.

for it appcars same as would

reciprocity motion at found at

th sohition we will take first the case in which discussing there is no friction. Tfjc coenieicnt is is then zero while a. Thc rca.1 part of th solution, rca. aud equal to p correspondis found by simply inb to th force .Fcos~, putting cos~)< for in (8), but it sccms scarcely nccessary to write th quations again for the sal of so small a Th same rcmark to change. applies the forced motion given in terms of y. appears is isochronous string, unicss it is not easy of given a. s)nall It that with th thc one motion of th beco'mcs natural infinite vibrations in case in the force entire

of the

point

to arrange

of application that a string

be a node; but shall be subjcct

practice to a force

th best method would be to attach magnitude. Perhaps mass of iron, attractcd perIodicaUy by an elcctro-magnet, whose coils are travcrscd But unless currcnt. by an intermittent some means of compensation wcre deviscd, the mass would have to bc vcry phc:).tion. to the imposition approximation may he obtained an obligatory A massive motion. fork of low pitch, cxcited a bow or sustained in permanent by electro-magnetism, operation of the exccutcs its vibrations in approximate independcnce actions order of any tbei-cforc light bodies to subjcct which any may be connecte(l of a string to with an it. point obligatory A better of by reIn small in order to avoid its Iiiertia Introducing & new com-

Donlnu'3 ~co)M<<M, p. 121.

154

TRANSVERSE

VIBRATIONS

OF

STRINGS.

[133.

traverse it is only motion. to attach it to th necessary extremity of one prong of such a whose fork, plane of vibration is perpendicular to the length of th string. This method of cxhihiting th forced vibrations of a string appels to hve beou first used by Meldc. Another hetter arrangement, for aurai adapted observation, bas been employcd Helmholtz. by Tj~ end of th stalk of a set into powcrfui tuning-fork, vibration with a bow, or othenvise is pressed th string. against It is advisable to ~e the surface, which cornes into contact ~ith t),e string, into a suitable (.saddieform, tho botter to prevcut shaped) and jarring. slipping to (5) we sec that, if sin X& Referring th motion vanished, to this quation) (according would hecome which Infinit, may be taken to prove that in thc case the motion eontempiated, would become real!y great-so grcat tl.at corrections, insi~upreviousiy ficant, rise into importance. Now sin when the force vanishes, is isochronous with one of thc natural vibrations of th first part of tho string, to be )tdd nxed at 0 and b. supposed of a ~onochord, string or other instrument with properly providcd a sound-board, it is casy to find by tnal th places of maximum rsonance. A very slight on eitlier side entails displacement a considerable o~In~e falling volume of tlie sound. Th points thus determined~i~ the a string into of parts, of length that the natural note of any one of them nxed at both ~hen ends) is tlie same as th note of th as fcrk, be verified, may readily The important of applications resonance .vhieh Helmholtz lias made to a tone from extraneous purify simple will occupy accompaniment our attention later, 134. o have Returning to extract Beie. now to the th real parts from occur as Thus let When a fork is placed on th

case (5), (R), (~ of

complex, 133. For reduced to

~f~~T~ the form

134.]

FRICTION

PROPORTIONAL

TO

VELOCITY.

155

corresponding By a similar

to thc

obligatory from

motion (8) 133, if

=~y cos~

at

process

to tho correspondmg obtaiu tlie forms of The values

impresscd ex, &c.

forco

.Fcos~

at b.

It remains

to

of a and /3 are dotorminecl

by

while

This If Hcd.

completes thc For

th

solution. be very small, the expressions may be in this case, to a sufEcicut approximation, simpli-

friction instance,

15G so tha.t

TRANSVERSE

VIBRATIONS

0F motion 0 ;ut.t

STRINGS. at & ?/ =-yeosp!; is,

[L34. tLc

to tlie obligatory corrcsp~nding of :uu!j~ bt'i~c~n nmphiudt''

)tpp~Xit)jft.tu!y

-which point

bccumcs

grcat,

but

not

inanit,

wheu

sin

= 0, or

thc

bo ])ot exprcsscd force, or motion, hy a single harmonie into such. term, it must first bc rcsolvcd Thc preceding solution to each componcnt may then be applicd and separately, thc resuits addcd togcther. T!ic extension to th case of more than one point of application of thc imprcssed forces is atso obvions. To obtain tho most gcnera) solution the thc s~tisf'ying conditions for the i)fitur:d vibrations must also Le addcd expression bnt thse become reducecl to Insignincancc after tnc motion lias been in progress for a suficiunt timc. Th !n.w of friction !msumcd in the prcccding is investigation th only one whoso resuits can bc ca.si)y fuilowed and doductivety, it is sunicient to givc a gnerai idca. of t))C effects of dissipativc forces on tlic motion of a string. But in other respects th conclusions drawn from It a nctitioua possoss on simplicity, dcpcndinr'the fact that 7'tl)e frictinn functionis similar in form to 7' which makcsthe normal of cach other. co-ordinatcsindepcndent In ahnost any other case (for oxample, when but a sit)g)c point of the string is rctardcd tttcrcarc no nonnfd by friction) co-ordinates so called. Tho-c exist itutocd propcriy ctcmcntary types of vibration into which the motion bc rosolved, a)id which may arc perfectly indcpendcnt, but thse are essentially different in character from thosc with which wc have hecn conccrncd for hithcrto the varions of the system parts (as affected by onc dcnicntary arc not simu!t!).neous!y in thc samc vibration) cases phase. Spcial no lincar transformation of th eo-ordinatcs cxcepted, real (with can rcduce and F togcther to a sun) of coefficients) T, squares. If wc suppose that tho striug lias no itx.'rtia, so that ~==() -~and F may tbcn be reduced to sums of squares. This probfem is of no acoustical but it is Intercsting importance, as bcing to that of thc conduction mathcmaticaMy analogous :utd radiation of lipat in a bar whuse ends arc maintaiucd at a cojtstaut
tem-

of application If thc hnposed

is a node.

prature.

135.]

EXTREMITIES

SUBJECT

TO

YIELDING.

157

Thus = 0 and .c = c:),nnot in whut hy

wiU

135.

far

wc

have

ttt string bc fLttu-uibd m prit-cticc, ma.nncr t))Ci vibrations cf t.bc points for

that supposcd is hcld at rest. it is ~ot without

at two Since iuterest

fixcd absolute

points, Hxtty

a yichHr)~
fm'tush

occasion

For

t))C sahc

synunctrical in is a-ttachcd cactt cxtrundty to a mass as uncxtendcd (trcatcd thc position of cqni(~t) towards spacc), and is urgcd by a spring iibrium. tf uo frictionat forces motion is nccessa.rity act, th
rcsolvabic iuto normal vibrations. Assume

of simplicity witti rcfurcncc

nf a string are liable of attactuncnt; and tlie prob!cm onc or two remarks of importance. is wc shaU suppose that thc System to tho centre of thc string, a.rid that

to inquire tu bo modiried

~=
Tho conditions

(~ sin )Ha;+~3 at th ends

cos MM'} cos (wa~ that

e).(l).

arc

whiehgivc

two

cquaitons, titc

suficicnt lattur

to dtermine ru.Lio, we Hud

and

tlic

ratio

of~S to a.

E)inun:tt.i)tg

(3) Equation fnund by writing

has tan

an

of adding togcthcr with its two arbitrary constants et and c, is necGssfu'ity th most of winch th prublem is capable, and is thercforc guncralsoiution to rcprcsunt th motion disduc to an arbitrary initial adquate of dispiacemeut tribution and velocity. Wc infcr tbn.t any function of x may bc cxpanded bctwucn x = and a;=~ in a-scrics of terms ~,(~,sin~)?~ H~, &c. bclug cos?)!) thc roots + ~~(~s!nm.c+cos~) of (~) and &c.thc + (5),

of roots, which may ho so tliat tan ?/~ = tan 2~, and the result for th corrcspouding each sohttions, particular infinite number

coi'rcspouding

158

TRANSVERSE

VIBRATIONS

OF

STRINGS.

values

~p~"

arc

of' t] i syst.em, From thc distant twoends,wherc~=0~d. as may bc proved T!.e Mncdc of thc Mergy l. also from energy string, y=~p) (4-).
thc ~I.ole nation Is made

co-ordinates

symmctry

of

thc

system "y Hcncc

it

follows thc

that s.,nc P~.

in each at points th =+1 1,

~sm.t-co~ 8 4

y of and that r; {S

up

of thc

of the SUl

masses

T!u~

+ cos M!a-)~ f~

+~+~+.r+~~(~sin~+cos~~+. Buthy theenaracteristic of normal property cannot ~lypr~eut sion for l' 80 that ~I~~tr'~ 1 pf" p~ + +

co-ordinates, in t..

terms c.prc.s-

sin 9?bX + cos ?M~) +,coq (~ (vr .9ili ili,x) (~ sin M~.t- + ces M,?;) (~ sin m~ + cos 7~) (~ sin M/ + ces M/) = 0. (G),

if 7' and g Le differcjit. This

how to dtermine s~gcsts th cont~orem arbitrary sothatthe stMts senc-s (5) au arbitraryfunction mayrcprcscnt Takc th y. expression y(.sin ~~+cos~)~.+~+ in it th type + cos ~) (~ sin ?x,~ (~ s; cos ~~) + eos 7. = Hencc scrics (.) sin exprc.ssi.,g

p~ .~d

cos Th

~(7) rc.nit is a

scries ~titutc of tcrms

of thc (~ sin +

p~~ + ail of

(., sin 7~? + cos ,)

vanish

whi~ js equal to th

hy (6), cxccpt thc onc for whieh expression (7), dividcd by

p~.sin~+cos~+~+~ and even but thus th coc~icnts bo rinitc, cftho thc series arc detcrnnnod. So nu,ch If ~=0

althc.g~ thc unrcstnctcd

prob~

p, is Instn,ctive.

strc.~

135.]

FOURIER'S

THEOREM.

159

on special and proofs of Founer's tho st~doit M apt, <,o ~cquirc' tor* contra<;tcd ofthosc rcsults important ofanaly~Is. We shall can bc deducpd if /t=
tcrniining

often

laid

sries, tliat Lapiacc's a vic~v of thc na-torc

now shew from

bow

Fouricr's

thcorom

in

our of th
tan

prsent string
M~= 0,

-X), the ends

?~ becomcs case

investigation. arc fast, and

or m~ = ~Tr, as

its ~encrai Let ~=0;

th we quation k~ow It

form thcn
demust

bu.

lu

this

t!)o sries

for y becomes

which

bc gnerai to reprcscntn.ny cnongh arbitra-ryfunctions at 0 an'l ?, betwccn thosc Innits. But now of.K, vanisiting suppose is zro, ~/8ti!l Thc ends of thc string th~t~ v!t,nishmg. may of slidiug on two smaotit m:ls perpendicuiMsupposed capable its length, Thc Mid the
tcrmina.l

must

be to

condition same OM ~e/~rc; vauish

is the and wc

vanishing Ic~rn. that

of any can

cquf~tion in fnnction y' whose be expanded

is th rates

of variation

a.t a? =0

and a?=

In a scriua

sign of a; is changcd, and th first series its without mcrcly changes altcring sign numcnc:il If tlierofore of x, magnitude. y' ho an even function it n'om to + And in the samc wa.y, if y bc (10) represciits au odd funetion of x, (9) roprescuts it betwecn thc samc limits. when th funetion of a; ~) (:r) may bo, it can bc divided Now, whatcvcr into two parts, one of w!iich is even, and the other odd, thus

Tbis

series

remains

unan'cctcd

so tha.t, if (x) be such tha,t it eau bc rcpresetitcd betwccn

(= (+ Z) and < (- ?) = ~)' (+ ~), th limits the inixed sries by

This possess

series thc

is penodic, with samo property, no

tlie

2?. pcnod matter what in

If

thcrcforc othcr respects

(x) its

160

TRANSVERSE

VIBRATIONS

OF

STRINGS.

[135.
This is

charactcr Founcr's Wc thc

may be, thcorcm*. now procccd in tlic

the

series

is its

complete

quivalent.

to cxa.minc

titc

c~ccts

supports, ~xcd. Titc Wc -whcn grt.

quantity shn)l conHne

cfu-:c (.)f :). striog whosc Lu g)-(.),t, cit!)c;)- through tnay oursulvos to thu two principal vanishes, (2) wlicn

of a slight yielding onds are approximat~'ty

of

is ~rea.L aud

vauishes

or throngj) cases, (1) and is

and

the

quation

in

is approxhnatdy

and To this a harmonie Tho in have effect tlie of approximation thc tones do not cease to form scalc, lmt tlie pitch of tlie whule is slightiy lowered is in fact thc same as that of an increase oftftcyiciding of th string in the ratio 1 1 + as might ordur

length

been Th

anticipatcd. is otherwisc if /t vanish, while Al is great. Hcre

rcsult

and Hcnce

Th

cfTcct is thus

cquivfileut

to a dccrcMc

in l in tlie

ratio

Tho tio~ is ro-outraut

~t au

for provi~g System' cmUess chain ~tchod of uir eucluecd

Fonrier's round

t]iMrem s,u.~th tube

from cyiiudcr

dynamic~I (S M!)),

considra. or thiD

culumn

iu a nitH.sIfnpcd

135.] ]
eonscqucntiy thc lowcr grcatcr and thcrc tho

FINITELOAD.
1. a nso in pitch, tone. Tt t]ie rise bcing

l(j~1
th

bc thought couiponcut nngl.b tttfit any kind wou)(t deprcss thc ptieti of thc string, ofyiL-!di!)g but thc preceding sficws that tins is not tiie case. uivestigation Whet))cr thc pitch will be raiscd or lowcrcd, on thc dcponds and this agam sign of on whcthcr tlic nn-tura.! note of dcpcnds thc mas-s I.s lowcr or !iig]tcr th:ui th~t of urgcd by t!)c spriug thc component vihra.tion In (question. of an ot))crwise proDcni unifonn string c~n'ying a nuite load ~at .;= ciui ho .sutvcd by thc formut InvcstigiLtett 13:}. if thc force 7''cus~< !n Fur, be duc to the raction against accuIcraHon of thc mass 136. Thc

which possible

comhined vaincs

with of

quation (or p r/),

(7) of

13~ gives,

to determine

thc

Thc

vfihtc

of y for any

normal

vibration

corrcsponding

to

is

whcre It ponents .ectcd

P :uid e arc fu-bitrfu'y docs not

constants.

wei~hted rcin~in unchanged, \vhi)c a)i thc odd components arc dcpresscd in pitch. hc takcn Advantagc m!Ly somctitnc-s of t)tis effect of a. wi)cu it is load, desircd for anypurp~cto distnrb the harmonie relation of thc comptjncnt tones. titG gravest voy g-reat, is wideiy component separatcd in pitc)i from !i]I t).u others. We will take thc case when t)te !oad is at thc cfnt rc, so t.hat = b = U.l. Thc quation in t])cn hceomos If bc

tn provc rcfjuh-c thn-t any normal anaiy.si.s cojnwhich have a, no.h at t!)C pnint of attachment are mitlie of tlie load. hy prsence For Instance, if a stnng be at the centre, its componcnt vibrations of evcn orders

th where 3/, dcnnting th.' )oad, is a sma]) quantity K.

ratio

of t)te may

masses bc

which

nf the stritig and caltud Th<~ <ir~ 1

162 root

TRANSVERSE to tlie

VIBRATIONS tonc of lowest

0F pitch

STRINGS. occurs v'hcn

~:3G. ~~is

corresponding sma,!l,andsucht)i~t

Thc obLn.incd thc hecn ta.incd ~:u'ts

second

term

constitutes chn.ptcr Thit.t th

a correction

to

tho

in a previous

st.ring

:dtogcthcr. at)d indecd cxpcctm), frorn thc string

( 52), by ncglecting it won)d bc i~htitivc formuia. a.s it

stands

vn.luc rcngh th(; Ino-tin. of might.

jua.y bc

I)n.vc
ob-

considcra.tion

that

in thc actutd

vibration

tlie to

two hc -\v!th-

of tlie

<'xnct,Iy so in <'ut C!)t)ufi!)g

in-c ncn.riy titc conipnting

On fmy apprcci:d)Ic <J)i.s s~ppu.sition thc rctcntion of t))e incrti:t of thc string incrca.scs thc kinctic to {t givcn vclocity of t))C Jond in cnf'rgy corresponding thc mtio cf whic)) icads to thc nhovc rcs)dt. This ~)7'+ mothod t):is indced thc In onc aa it )ni'd)t bc a.dvantagc rcspuct, is

str&ight, and may bc n.ssmncd kitK'tic n.nd potcntifd nergies, crror in t)tc cn.lcoh~tcd pcriod.

is not nnifortn, or ncarly uniform. ~)] th:tt whcn is t.ha). ),hc )oad -/)/ shoufd he su(H(.-icnt)y prcdouioant. ncccss.iry npplicd

13C.]
Thcrc thc the pairs; mn.tc!y Thc two The of a small 137. second load. for by is no

CORRECTION

FOR

RIGIDITY.

1G3

othcr

root of thc aftcr

of ('t),

until

component T!ie roots oue set

.string,a, thc first occur hy ~X~==S7r, in which tho

sin~X~=0, vibration ni ~nd

gives of indcpcndent

which

is givcn &'7T./)/

closely contiguous tho other approxitcrm is sma.

~=N7r+-

second

types
goncral

of vibration
formula

for N = 1 are

shcwn

in th

Hgurc. the cifcct

load on thc Actua.1

to find (2) may a)so be applicd components. pitch of the various wh'es arc not

Thcy oppose into two pa.rts,p)'oducing and shcws itself viscosity, produces servative no sensible

and strings a. certain rsistance two

to bcnding, distinct enccts.

flexible. perfectly which may bc divided The first is called

th hy df~nping efTcct on th poriods.

This part vibrations. The second is conto the thc potcntia.1 cnorgy A eom-

an<t contributes Iti its chtu'a.ctcr, of thc system, with thc effect of shortening

pcriods. bc givcn cannot hcro, but thc phjte convcnioltiy investigation case 'which is most intcrcsti))g to musical instmin its application adinits of a sufficicntly mcnts, simple treatmeut. Whcn specined vanisties. (1) (2)
case f/.C'

is takcn intn rigifhty to thc with respect Two cases Mrlicii Whcn Whcn
= 0.

account, terminal

somctiling conditions noted

more tha.u

must that

ho y

may

hc particularly

tii(, ends t)ic eiids thc

termina)

arc clamped, are dircftions

so that tli~it are

q = 0 at tt thc tll(" ends. C.,n(ls. = 0 free, in which

pcrfcctiy

Itis

th

laLLct'whichwc wo'c
y

Jf tho'c

no ngi'tity,
,c~r,r, p

propose nowtocnnsxtcr. t)tc t.ypc of vibration

f satt.sfymg '1 t!ic second

wouht

hc

1 cnn'Ution. 1,

si~L

Thc hnt from the thc

cicctofthc such

whethcr

ri~iditymight a rcsult occur

bc slighUyto or not, thc on thc corr'-ct

distnrh priod

the type; calcntatud

type ot'dcr of.stna)) N')\v

a!)d kinctic nergies potcntiiU rc)n:uns Is nccc.ssarily mudtcrcd qu:)nLit.ic.s ( US). duc to
thc

that supposition as f:n' ~s thc first

Dit' potc))ti:d

pocrgy

stiincs.s

is

expresse'!

by

~64

TRANSVERSE

VIBRATIONS

OF

STRINCS.

[137.

/e is a quantity n.nd on t.he form ofthn prepai .;(i required

amount K, u.nh.j.

where

dcpcnding
~cticu Ti.e/u. in a

on ur~'

the

nature thn.t.

of thc bcc-msc

mntcrial ttic f~i-cc to thc whoic in ds to

pir.infr

nrc. nnt. ncw and Thc 1 p

is vident,

to bcnd of bcuding w))icli mu.st

any

work

is thcrcforc which

Is proportion. to c-iTcctcd, t!)at is to a]rcady bo donc to producc a curvaturc to ~-p'; whl)c 1 = 1 and M.<, p to
thc

clment

proportional =

app)-o.ximatnm

we work

if T. dnote what tl)c pcriod wouM bccomc if t])c st.nng wcrc cndowcd with pc.rfect It nppears that t).c cffcct of thc fi~ibiHty. .st~ncss n.crc.ascs wilh t].o ordcr of thc rapidiy vibracouinent ttons, which cea.sc to bc)o.~ to !L I~nnnnic scalc. i.t t!.c Ho\vcvu! in niu.sic, t).c ton.si.m Htrn~s cmpjuyc..) i.s usn~Dy suficicnt to rcducc thu inHucnce to
ofrigi()it,y insignifiance.

T))G )neH)od offhis section cannot hc without inodin~ppjicd c.~ion to t).c ot).e.- case of t~nina) comiition, n:unc)y, whcu t).c cnd.s arc c-hunpcd. In thcir immudiatc t)ie type of nci~hbuurhood vibration must dm' from that a.ssuincd Hcxible by a po~cHy stnng w),.d. is no !o)~c.r byaquantity, s,n:dt, and w).osc square thcrcforc cannot be nog]~t~d. Wc sha)) rcturn to this suhject,
wttcn ti-cattog of thc transvcrsc vibrations of rods.

J38. TLct-G i.s onc p.-obicm <o t).u vihratiut.s rdating of.strin.s whtdi wc I.ave not but which yut considcrcd, 1~ of .s.~nc practi~i intcrcst, thc cluu-acter na.ndy, of th nation of a vioiin (or ccHo) undor thc action stnng of thc bow. In this prob]e.n th ~o~s oft!)c bow is not ~W!~ u).()c.stood to aUow us to .sufHcicntJy foHow cxeh.sivdy thc M ;)~~ mcthod thc indications of thoory mo.st bc .supp)emuntL~ observation. a ).y spccia) dextc-rou.s By combitiationof cvidcncedrawn frorn both Kourcc.s !Id)nho!tz)tas -snccccdcd m d.r.nining th principe tcaturus of thc. cas~ but somc of thc dtails arc .stii) obscure.

138.]

VIOLIN

STRING.

165

Since wc infcr strict

thu thaL

note

of a. ~ood Lh~ yibraLiuurt

Instrumr'.nt, a.i'c stricHy

or qnitc, tite sn.me vibra.ttons, although iti sutuc sens frcb. ibrccd, arc thus Thcy are whony dpendent for tbcn' mn.intcnn.ncc on th energy drf),wn from th bow, and yet tho how doos not dctcrniine, or cvcn sensibjy mod)fy,their pcriods. We arc rcmindcd of thc scif-aeting clectricaL whosc intcrrnptcr, nctu'Iy, TIic motion
offrcedom

is thc td'td. periodieity antt!ic note clicitcd by the bow lias note of th string. pitch a.s tho n:itu)':d

well ha.ndlcd, is musicn], or at least that pcriodiu, Morcoverand this is very import-

is Indcd
which

furccd
consi.sts

in thc
in

tochnica.!

sense,

but

haa

t!ia.t

kind undci'

dcturjnitnng

(who))y,or

in part)

what

influences But it
witti

it sha,ll not at

coinc. once
pcriods,

docs
its

fullow

from that it

th

fuct

thn-t

tho

vibrtes

na.tura.1

confortns expansion

to its

string naturnl

types.

If thc

coefHcients

of tlie

Fourier

co-ordhiatcs independcnt by wlticb thc conngura.tion oftiie is at any moment we kuow that whcn de~ned, system oc tliere is no friction, or friction such tliat titc na.tur:U vibra.tiu)is arc cxpresscd n. s~e cach co-ordin:tte harmonie by ma.king Hmction of th timc; h:m (or quasi-harmonie) while, for a.l). that hitticrto to t))e contrary, eacii co-ordin.~to in the prsent appeitred c:mc nii~ht bc M?t~/function of tim time periodic Httle examiua.tion will show that tlic vibrations in their typos as wcti as in thcir periods. natural Tho force excrciscd by by the bow at its point in time'T. must But a hc sci)sib)y

be takcn

as

tlie

of application

may

bc exprcsscd

so tha.t

tlie equation

of motion

for tlie co-ordin~tc

is

& being tion,

thc pouit of appHcn.tion. will give a corrcspondin~ tcrm but tbe ono whosc

Each

of the componcnt of its own pci-tod in

parts of th solu-

as tho natuml period is th same po-Iod will risc relative of~ cnoi'tnousiyin then, importance. Pra.ctienUy if tlic damping bo suialt, wc uccd tha,t p:~rt of only rctain

166

TRANSVERSE

VIBRATIONS

0F

STRINGS.

[138.

W whicL 11(:"

( 1. S 011 d' j"!en! pends on

J, ij.

c, c.~ i" e~ ~

thut ) tua!'

I., tu ci

~y, s.ty,

w~ Wu

may ~n.y

rogm-d regM'

the vibrations Anuther

as natural matent fact,

in their

types.

drawn both from supported by cvidncc and aurai is tins. theory observation, AH component vibrations are absent which have a node at the of excitation. In point to extingui.sh ordcr, however, thse t!)at the tones, it is neccssfa-y coincidence of the point of application of the bow with tho nndc shonid bo vcry c.mc<. small dviation A very t)tc rcproduces tones with considorabJc jnis.si)ig strcnf-tb' Tlic rcsts, rcm~inder vas dcrived As of tho from vidence direct on w)nch He!mboltx' with thc theory vibration-

observation

in Chapter explained curve th rcprcscnting a~ it would e seen traced observation, In ordcr to Jcduec parent cylinder. microscope. aview oftbe its ordinary be uni-otiud, The

affurds n., t)tis instrament motion of thc point undcr on tlic surface of a transt!te reprsentative curve in must be conceived tu

form, the imaginary cylinder or developed, into a. p)anc.

results arc obtaincd simp]cst whcn the bow is a.ppl!cd at a nodc of one of tho higher and th point obscrved is componcnts, onc of the otlier nodes of the samc If th bow works system. fairly so as to draw out tho fundamcntal tono cicarly and strongly, thc reprsentative curve is tha.t she\vn in figure 22; where tho abscisse to tho tirne a conpicte correspond (~173 hcing period), and t]io ordinates the displaccment. The rcmarkabio reprcsent

fact

is discloscd

t))at. thc r-T., during is consent,;

whotc

parts ~ nnd sct-vcd point goict'a! Wc terms.

T ma.y bc divulcd po-iod c~ch of whieh thcvdocityof th vulucitics to and fro

inin ;).re

two in

thet)b-

but

uncqua]. Iiavc If now t)tc to rcprc.scnt of timc ori~in this curvc by n so-ies ofiiarmnnic to t))c point J, and
13).

corrospond p.

Donkin'<~cf)~.<f)~,

138.]
J 7' T'Y?== Fourier's

VIOLIN

STRING.

1GT

t)teorcm

E;ivea

With pouents

respect of~ must

to thc vauish

value for

of T., wc know which sm-=0

that

ail (xa

those being

comthc

th circumstances 'bccn.usc under point of observation), is thcreforc them. Tliere the bow cannot gcnGrato thn.t T. T= a', l; and in fact observation suppose J.C' which C~ (iti tho
tlic string

of the case reason proves two parts to tbat iuto

figure) is cqual to the ratio of the is divided by the point of observation. vibrations of thc string are

Now geucral by

thc

i'rec

rcprcsentcd

in

= sia and Uns at thc cos agrec must point a; = we inay write

+ with

sin (1). For convenience

of euu)parison,

2S7T< 287rt 2S7T< 2S7T/. = A Co. ces 2s7r < cos R sin Sin ~~7rt + B~ C~ (7, ~, T T T t D ~(<), and it thon We find a.ppears that C, = 0. D,

T~ ~) 2/ 19 1'0)

Ir

a.Iso to dtermine

whcucc

ca.se reserved, the !c:wcs DH undetcrmincd, comparison but wc know ou otlier groundH tliat DH then vanistics. Howcver, for the sakc of simplicity, that. for tlie pt'cscnt we sh:dl suppose of tlie bow If the point of application D~ is ahvitys givcn by (2). do not concide thc with a nodc of any of the lowcr componcuts, error comtnittcd On tliis will bc of nu grcat tlie cousof~uenc. solution ofthc problem is

In thc

undGrsta.uding

complote

168

TRANS

VERSE

VIBRATIONS

OF

STRINGS.

['138.

The

a.mpl:tudea

of th componcnts y

fu-e t.tjercfore

proportions) A If J Iftijc v~nish,

strincr. t))e fun~oftwo m

to 0 string but thc

T))c

int~u!p!!t~,)dMt!-l;

~)U;tdfu!fi'JC<;)t'C!t)~)"' 1 l L, whic)i is th
)!tw t]):tt thc as

futictiou i'uuctiou s'sin'S~ s-"sln' ho

Oth! c()ua,tin .st.mi~ht

sonicwhat l, cvcn
obta.hjs is

'J sinnJfn-.

ptucked
oncs

at foHow
(3)

th
thc

mnhilc,
same

components
fur :Jways a vioiin

ulcHcatcs

strmg

a.t an angtc. mcetittg to shew this, !ct us thc origin change so that mu)t,ip!ier,

Unes

In

order

inore a.ud

convettiontiv
t)ic constat

of tftc tune,

will hc thc Now

quation

cxprcssing ( 127) thc fixod

that

th
thc

form

oft))c
of

string th and

:tt any p.ur

titnc.

wo know

C(tU!ttio.i

of lines at a

from proceeding puint wliosc co-ordinates

em]s

of thc

.string,

arc or, /9, is

mcet.nrr

Thse cf thc forwanis .suction '~t' whieli chofti. Since

quations of point bctwecn itscif thc

i.s

indicatc intersection .~=0 an.!

situatc.d

that

thc

projection

on th

axis

of:B

equilibrhun motion

backwards and uniforndy an.) t.)iat t).c point ~=Z, of interon onc or ot).< uf t~-o p;u-abo]ic arcs, of th .string positon is a connnon

moves

as th~ d~nu<i string of thc hy tl.at of point intersoction of its two bas no cspccia! straight parts, rctatioti of observation), to (Lhc point it. fo)h~-s that, accordin. to t])c.se ouations, titc sa.ne J<ind of motion a't m.gbt Le obscrved othoAnd t)iis is any point. trnc. But tbc thcoapproximatciy rctica) rL-.suk, it wil! bo was o)i)y obtaincd romonbcrud, by asin certain H)))ni))g tbc prsence proportions vibrations ofcomponent nodc.s at ha\'in~ in tact thuir abscucc is )-<j(p)ircd by thongh "chanica) !aws. Thc prsence or absence of thse components is

the

ofthc

138.] a mattcr but not vibration

thc

STRINGS

STRETCHED when a,

ON

CURVED is thc

SURFACES.

109

ofindifercncG in nny curvc

conponcnts

of observation, point i.-i doparted Wh~n thc nid.' from, thc otho' c-f. of duc to thc absence of ripples, a sries shews will be fuund furt!icr dctails Somc In node
question.

in Hcitnhoitz Thc

and

Donhin.

of thc bow dcpeuds upon the fact that sustaining powcr so thf).t is Ic.ss at modra te t))an at smalL velocitics, so)id frictiuu witil th bow is actud of th stri!~ moving w)tt] t)tc part npou action is greater thc mutual at thc s:~mc vulocity), (nut imprububly direction with in tho opposite is thc string moving titmi wi'eu eH'cct in tl~c first Th relative ~ccctcrating :L greatcr vulucity. ncutratiscd is thus not cntirdy of thc motion by th subpart accctcra.tion rcmains an and outstanding rctfu-da.tion, squent losscs of in spite of other th vibration cap:(.b)c of ]n:untaining of solid friction samc peculiarity A cm-ious cncct ofthc ~ncr"-y. that tl)0 vibrations who found bas becn obscrved by Mr Froudc, or bc maintained from a sbaft might uf a, pcndulnm swinging shaft to rotate. tt~c cvcn Incrcascd by causing will in cui'vcd surface on a. stnooth stretched A strin"' to certain conlie along a gcodcsic Une, and, subject cquilibrium if distilis about will vibrato eonnguratiun, of stubitity, ditions is when. tlic call bc proposcd case that TI)C simplest ptaced. and th cquitibrium of any form, is a cyHnder position surface Th studcnt hncs. to titc gnerating is perpcndicular of tlie string of thc curvature is indcpcndcnt will casUy provc that tbc n~otion arc in :dl essential respects and that thc vibrations of tlie cylinder, Th case into a plane. wcrc developcd thc samc as if th surface round th eylindcr, is a nccidace endiess of an funning string, 139. v/orthy of notice. tlic charactcristic thc
surface

In oi'tter to Ulustratc we will tako problenis,

string v/hcn in stretched cquilibrium, on th

features simple
smouth sphre, Tite

of this cxample and

co-ordinatea

class of

of a

conparatively
of a

which Jatitudc
~n~itudc

it

circle. :dong a grcat to refer convcnicnt will be most circle thc grcat front mcasurcd
measured

lying, to are th

thc

system

alung

it.

If thc

and th as equator, radius of thc sphre be

wb bave

170 Thc

TRANSVERSE extension of th

VIBRATIONS string is denoted

0F by

STRINGS.

fl39.

J(~i)~. Now
~=(~f~+(ocos~

so tliat sothat f~ = f/ {(~~ Thus -8~(y-(~' a.nd li and V= aTl. ~)~. ends Le fixed, + e Jcp. Cltp dtp.(2);1 = 1/ 2 (~~

~PP'tc]y.

i-~ -10 Q

ose ~(to

If thc

~=0'L~J. and thc equation of virtual 0 se dtp 0 8~

0 is + 8 dcfJ = = o, 0,

velocities

0 whence, since

o0

(10

S~ is a-rbitrary,

This is thc

cquatiou

"(~) of motion. oc cos~<, _rl'B ~,+~0. wc get 22 (4), to t)ic condition that vanishes

If wc assume

0 cc'p subject

cf \vl)ieh with is

the

solution,

~=~sinj~~+l~
Tho rcmaining

.cos~ is that

.(5). vanislics whcn

condition

to bc satisfied K.

~ =

or <j& = et, if a = <! givcs

Tiiis

~h'
1) ~herc a ?~ is an iutcger.

I\ p a

-1 -~=p'(~p

( :I- cG~

.G

CambrMHOMathcmaticftt TritMB Exnmination,

187G.

139.]
Tho
stnufht

VARIABLE

DENSITY.

171

normal
strmf. viz.

functions

arc

thus

of

~'c

samc

form

a,a for

a.

but is to

thc series makc

tonc of a thao. the corresponding graver of p2 is ngaof tho values If a> 7r, 0110 at least string. straigbt If are unstable. modes the corrcspouding tha.t tive, mdica.ting iu tlie disis zro, tlie string a == '7r, bcing of tlie same length as whe!i = 0. position, placed A similar a striug tion. 140. method strctched might be applied round tlie equator to catculatc of any surface the motion of rvoluof

of periods cach tone

is digrent.

Thc

effect

of th

curvature

for a vibrating of the problem solution approximate bas been uniform dcnsity longitudinal string of ncarly but not quitc of in Chapter considcred cxampic IV. 01, as a convenierit fully It will bc of approximately thc general simple systems. thcory thc bc ~+ If tlie density hre to repeat thc result. sufficient The pcriod , of thc ?' component vibration is given by

Thse
small

values

of r" arc 8p and slight

correct

as

far

as thc

first

power

of thc a. coralways at

quantifies

rection occur

for such iu practice.

give the incans of calcul~ting as must from uniformity dcpartures ?~, and of a small tlie points mcrcly load

As might be and rises nodes, conscutive Gstimato

formula

th effect expecte(l, at to a maximum WIien effective it

vanishes

nodes. of thc
indicatcs

is dcsircd

midway to make

bctw<;cu

hood

of loops 1-tl. The whose

uniform dcnsity of a ncarly to the is to Le given tliat attention rather than to that of nodcs. dinerential longitudinal quation dcnsity determining p is variable, th is

a rough thc string,

neighhour-

motion

of a

string,

172

TRANSVERSE

VIBRATIONS

0F

STRINGS.

F 141.

from

which, if .'<))';n:u iuucti~

wc

assume

oc cos

wc

obttuu

to dctcrmu~

th

of th second for~-?', This~uationis hncar, b,.thasnothit)icrtobccn.so!v.d:n o'rand huitotcrms. Cun.s.dcrc.t th curve ~su.ucd dcH.ling by tho .st.rin.r In thc uonna! mode uih!cr it dutcnnincs considration, thc c~rc at nnd any pou.t, cinbodic~ a ru)c I.y whidi accordin~jy th c-n-vo can bc construct.cd
~phic.Hy. Thns in thu

~yhcre

,swntt.n

string quation
jaay

nxcd and w.Lh

out

both
wit).

npptic~iun

to

cnd.s, if
zro

st.t to

from

inclination,
trace

curvatm-c,

what
th cutirc

eurvaturc
curvc.

nn ~rbitrary ~-0 are ahvay.s directcd by tbe and in tins way wc p.-uceed,

c.ithcr

end

the curvc will cross rigbt, axis of at thc rc.uircd aud thc ]aw of vibration distance, will hc con~tctdy dctcrndncd. If Le nul known, ditterent vaincs bc tri.d untii may thc curvc ends rightiy; a sufncient to tho approximation value cf~m.~u.su~iyho ~-nvcdathy~ c~cul~tion founded on an as.smncd type ( 88, 90). Whcthcr t!.c longitudinal be uniform or net density th timo pcncdic of any simple vibration varies c~~ as thc root cf thc s<(u~e aud den.sity us th .square root of the Invcr.s.ly tension undur w)nch t]io motion takcs piace. th. of prob)cm dct~mining and H,c type of vibration pcnod arc gi vcn, tins purposc is ody <o substitutc necessary and of its second di~brontial cocmcient in tbedcns.tybo th extrunutics innuitc, of zero curvature. W!tcn a
givcn string

If

thc

assumcd

value

of

be

Thc

eonvcrsc

<icnsity, w!,c.u th is always sutuhic For thc givcn vah.c of v Unkss quation (2). a string arc points of

thc

is s)iortencd, tono is every componcnt ra..scd ,n p.tc)L For tho new stato of as things may bc rcgarded dcnvcd from thc old by at t!ic proposed intradnction, of point of a spring hxturo, (without ~vhose stifFncss is inc.rtia), gradua]!y incr~scd without limit. At cac)..stc.p of thc proccss tho potcntia! ofa givcn dformation cncrgy is angmentcd, and t).c-rcfor ( 88) th intch of every tone is raiscd. In likc manner an addition to thc length ofa str.ng thc pitc! dcpresscs cven though thc added bc dcstitutc part ofiucrtia.

142.]
14-2.
Sh' ~c'

VARIABLE
Atthongh
<< pr,ui'

DENSITY.
of t'~

173
quation (2) of141
.uc'ft.h'

a gnerai
v'

Intgration

m.T\p~y

t!n)-~k'n 1.

of th hnuar of th solution quation intcresti! propcrties Stnrnt liavc hcen detnonstrated second order.which LyMM. tins work to give anythiug in It Is impossible and LiouviHu'. in bot :), sketch, of titeir invc.st.i~Lions; ~ccomit hkc n. compictu intcrcsttho te:tdi))~ fca.tm-cs n.ru inctudcd, which m~y be found comicctud with th on sone points li~ht incr, and will tin'uw bodics. 1 hve not of continuons thcory of the vibrntions gcncnd many of'thc thought
io thc

it

ticccs.s:u'y tncmon's.

to adhre

vcry

c~oscly thc )

to t)'c mcthods

adoptcd

origina.)

A.t no point

of t!t0 curvc

satisfying

cquation (1) diHcrcntin.tio~s simnitancousiy, &c.

C~:Cs

rl'r/ 2 ~+~~n.(D, can both y a.nd '( vanish prove togctiicr. that, If

By sucsstvc n,nd vanish

of (1) it is c~sy to aU thc highcr at th must

dncrcntial samc

coemcicnts
Cl'`

musta!so tho

vanish curve

eoincidc
\duc conc:wc

and tilo'cfore point, with th axis of a;. be ascnbcd

tbt'oughout

by T~ytor's

theorcm

Whatevcr
suu~ being

to

thc to\vard.s

cn)-vc tlio

satisfying axis of a-,

(1) is
sinec

is cvoywhcrc cverywllere p is thc Lu positive, bclow a curtain If remain Wc which Lct tion hve bc

Ilositive. positive. ordinatc limit will

If

at rcnmin

l th

origin

vanish, 1

ana and

~Lx

dei'cndcnt
cm-vaturc

thc vu'y smaH, on t]tc positive

for aU vaincs of a; positive to on thc vainc ascribed artd thc curve will is slight, thu axis distance. for a gi'cat aU tho vahtcs of a;
in magnitude.

sidc

of

now

satisfy

as that to provo incrcascs, diminish thc cotation = 0 gradua)Iy oi-dinatc of

a second curve

Le th

sati.sfying

th

qua-

+ ~+,=0 cl;c'

Il

p?J

.(2), at thc origin, and lut us

as weil suppose
T)to LiouyiUu's

as thc condition t)iat

)i)c;]))"i)'n .yuto'/t'~

that

vanishes grcatcr than

nrt:

is somcwhat
rnferrc~ ()'S!it!j. to ui

Multiplying
iu tito first

(2) hy y,
volume of

tho

tcxt

euittttmcd

174

TRANS

VERSE

VIBRATIONS

0F

STRINGS.

[143.

(1) by bctwecu ttic

wii,L.'t!,

n.nd

and integn).th]g subtmcting, limita 0 and x, wc obtn't), sincti

with and
.1

to x respect, ~ot!) vnmsh

If wc further
y vcmi.sfK's, ;u]d

suppose

thftt di~rcucc

.c

corrcspomiH

th~t

tlie

betwccn

we gct

to n. point :).t winch and is vcry small,

ultimately

TIic wo Jcaru w])ethor as that for which If betwccn

ont as th

right-hand that y and

mcmber arc of

of
thc

(4)
samc

bcing
sign,

csscntially and

positive, t!)at, sign of a;

thcrcforc samc

be positive to which
vanishcs

01- ngative,

is ah-eady

of thc

or in o~K-r words, th value y is changing, is less t)ian that for winch vanishcs. attention .r= thc
incrcases,

wo Hx our ;K=0
value

on

thc

and
of

ordinatc
until

of the curvc portion !yin~ contitmcs positive t)n'ou"'ha curtain v:duc is attaincd,

which with

wo will thc first

call
normal

Titc
function

at 0 and
iacreases,

unti), second value is attaincd, th curve special crosses th again axis at th point a'=~, and thcn rcprcscnts t]tc sccon(t norma! functiou function bas thus onc internat M,. This root, and onc In likc manucr t.o a hi~hcr ot~y. value wc corrcsponding ohtain titc third nonna! functiot ?~ with two interna! roots, and
so on.

lias no a~d thc first root

root
movc.s

is now idcntical function of a string of dcnsity M, As cxccpt at thosc points.

inwards

in form /) (ixod a'~ain when a

from

a;=~

Thc thc

?"' functiou dinurcntial it

sinco with

its ih'st

M,, bas thus cocfticient

sign

cxactiy ucvcr
cach

1 intcrnid roots, and vanis))cs sinndtancousty a root is passcd.

function, quation

changes

titnc if nnd

Frn)n normal

(3) it app~u-s

that

hc tw.) di~'rcnt

functions,

A bciUttifu! to from the mnnhur Hnitc

thcorum

bas routs of nurnud

bccn

di.scuvc-rcd

by

Shmn

rc)!~in<r

uf

Uic

ci'

funcLio;) If

(k'rivcd

by

addition

tiumbur

fuuctious.

bu thc eompoucnt

143.]
of lowest order,

STURM'S

THOREM.

175

and M~ thc

component

of highost

order,

tho functtou

whcrc internai

~), roots,

< and

&c.

arc ~os~

arbitrary Ml

coefHcicnts, intcrna.1 roots.

has

a<

~ecM<

m1

The roots

cxtrenutics in a.U cases. to that givcn not less

f~t ~=0 The

aud

following hy Liouville,
rigorous.

to .~=~ con'cspon.d of course dmonstration bca.rs somc rcsonbluncc at but is considcrn.bly simpicr, aud,

1 bellcvc~

Internai that /(.E) ~as cxact]y suppose number ofwincli may bo cq)ia.l), tho derived functionj~) Iiavo less tl)an mnst + 1 internai roots, sincc therc onc root of/'(.'c) bctwccn cach pair <~f consccntivc roots 'wc t.Itu whoc manncr,
bcsides

If

roots

(any cannot

bc at ]cast

numbcr wc sec that

of roots thcrc

of~(.~) must

eoncurnud bo at Icast

ofy(a;), and is ~.+2. 1); liko roots ofy'(a:),

tlie

to
that

roots;
any

which thcmselvcs cxtrenutics, necessarlly correspond so that in passmg from _/(~) to y"(~') it is impossible
can bc lost. Now

roots

bas

/t interna.! roots; and th proccss tnay bo continnc'd tu fmy uxtcttt. la this w~y wc obtai)i a scrics of' ftmctions, :t.]) intct'n:d roots at !en,st, whieit dUrur from the origina! with relative of imtCtioM/(:)') by tho continua]]y menjasin~ !))iport!incc the componL'uts of thc hi~Lcr oniurs. Wi~cn t!i(i procL'ss I):~s bcot
ciu'i'iud sufficicnt.Jy fur, we sh~H :),n'ivc !tt a function, whosu iorm

at

luast

ditturs

fonction uf p!t.'asc ft-om that of t))c normal )t 1 intcrn:d roots. ordcr, viz. M, and w)iic)i ])as thcnjforc hi~))cst IL funows no roots can 1je lest in passit)g down thc thi~t, sincc uf fu))ctio!)s, UXC')) ;) t. so'ics thc m))ub(.'r uf Int(.'t'))a) ruuis ufy(;<') c;U)n<jt

as )itt)c

:)H we

17G

TRANSVERS

VIBRATIONS

0F

STUINOS.

[142.

Thc

other

]~!f

of tho

thcorcm

is proved
h.kw.'n~

in
:;i

n..simDar
fl'u~i'

mannor
h. ll

hycc!tt:t~:u~th~8';)-Ic:~if'un.cti.< Utiswaywcobt:).i)i

arrivingnormal

t~t last fu])ctionof

at

functiun

sc.n.sib]y

coiuc.d~nt v~ ~nd

in form havin~

with

thc

ordcr,

Lhei-cforo Lave that

M-lnitcrnali-oots. Sinccnoroutsc.-mbctastinp~sin~upthc senesfrom <.his functinn it full~y.s to/(..), ~nr~ot t].at/(.r) fewertntcrn~ roots tl.nn ~-1; but it must bc und~-stood number oft)~ w 1 roots auy mny be cqu.d. Wc a!I tlic wi!! now prove cocfHcicnts (G) hy p cannot be ide..t:ca!!y tl.at,/(.) va.n.sh. Supposa t).at and intgre wit), respect to~

zcro un)c~ is not 7ro betwccn ~nnc t!tc

Muhiply Jhmts0aud/.

Thc.tby(5)

from

wl.ich,

since

sec t)iat/(.r)
t'fmgcofintogratio)).

t)ic intgre on thc rigbt-hand sidc i.s ~nitc we cannot vanish fui- aU vah.c.s of Incli.dud withuAhc ].M inadc u.sc of Stur.n'.s thco-ent to su as and

LIouviiIe

sliew a;=~.

f.u.ctions n~y be eo)np<,u)h)c<I at-bitrary sign atatt bL.twccn puint.s iymg ~=0 methodi.ssunK.'wItatasfoDuws. Thc &,c, suppose nants, va]u~ of~ fur windt th n.nc~iun

scriGS of normal

i)ow His

a an

to have

is to changr.

sign

bfi.)<

(.juantitic.sw).)) tu bc aU (tiOcrent,

wiDxmt loss fut u.scon.sidc.r

ofgcnera.Hty~G m~ tl~suries of dtermi-

~andbyStunn's T.c~.snsa!.nc..u-funct,onof.,(.)~.) t!K..o,-cm h.~th~forc onci.iU.rna! n,o< atn.ost,whidi roulis ]\rcover t).o cvidc.ht)y dL.tcrunnant is not idcntica))y zcro sn.ce th cu~ciont of ,(..), viz, .) ~.t ~)~ ~~), ,,]~tevcr bcth.v~ho ut' -\Vc hve thus oht.in.d a function, ~h:ch chan~s~natauarLiLrarypuiut.r/ui.!thcreon)yiuL.rn:t))y

143.]

EXPANSION

IN

SERIES

0F

NORMAL

FUNCTtONS.

1~7

The und,

second

dterminant hnvc

vanishes tnure
tb~n

when
tv.u

~;=~,

and
~oms,

witen
it

~=&,

~inc

it canuoL

iatcmai

chu-u~ca

t)ieso and Thc tl)ere on!y. values, sign, wheu x passer through coefHcient of ~(a;) is tlie value nssumod by the fii'st dcterminn.ut Hcnco thc secoud dcterminaut whcn x = &, and is thcrefoi'e finite.

is not

identically thc

zro. and series vanishes = c, and a.t those and wilcn of ~(;E) is funtc,bei!)gthc .E= c. we can form process whieh s!)all vanish functions, this in th

Simila.rly changes internai value It functions

third dctermma.nt = a, when a; = sign whcn x Thc coefficient points only. dterminant that wheu
Ing

of the second is evident

compounded values of a*, and not eisewhere aud change sign for any arbitrary whose sign or, in other words, we can form a function internally; is arbitrary 0 to x = over th wliote range froin..B= On possibility .c = by of the this theorem Liouville founds his function demonstration between of the x = 0 and

by continu of th normal

an arbitrary of representing a series of normal functions. and take

If we assume

the possibility

expansion

t!)c fiad

necessary

values 1

of < J

<

&c. arc

determiued f8l

by (9), a.nd

wc

/(.c)

2 j'~) (.

fp
~o

i').)~

p !t;(~) j -'a U

r~-}.(11).

If

tlie

sries

on

thc

right

bc dcuotud and ~(x:).

by ~'(.~),

it

ron.Lms

to

cst.abtisli Iftiie Intcgra.tud

thc idcutHy

of/(..t')

mcmbhr right-hmjd to with respect

of (11) he tunttiplicd by pi<~(~) and wc sec that from a:= 0 to x =

or, as we nmy

:tlso writc

it,

where

M~(-c) ts ct~y nonna.)

function.

Frutu

(12)

it follows

that

w~crc

the it.

coefHcicnts

&c. are !n'bitmry. ~),

178
Now so tuLs

TRANSVERSE

VIBRATIONS

0F

STRINGS.

[143.
bc pnssiMo

if F(~)-/(.c) bo not chonsc t.hc c<.tis~)t.s .1.

thc sa).tc sign

Idcntic~Hy t~ J,.

xcro,

it will

~'Y.'(.<:) -), eloncnt ofthe woul(! bc positivu, jutcgmt :m() cqu~~ion (13) cuu)d not bu tn.o. It fuUuw.s Lh.Lt F(.) cannot <)i~r imn) zro, -/(.<.) or th:),t t])e s(;i-ics of sonnt fnnciions tho right-hatxt fortning nimber of (11) is idcntie:d for ~11 vaJue.s of .t- from tc= 0 wit)i/(.r) = to Thc and rc.suits arguments tu thc ]):u-ticu).-u- case arc circuhu'.
of

throughout

<M -t .< .r~ in' \v])ic). CMc'cvc.-y

Uns

snctinn

arc of course stmig fur wi~ch

ptic-aDo normal functious

of a unifonn

~~ptite

14.3.

Whcn

Lhc vibmtions

pJanc, it is usua)]y ninst cxGcutc-d in perpendicuL-upcndonUy. Thcro notice p~sing tt'catcd witliout Suppose tha.t

of a string arc not con~icd to cno e.mvenient to rcsnivo thon into two sets which ~imc.s, indcm.~y be trcatcd onc
niotion

is, howcvcr, in which thu

case
i.s

of t!.i.s description worth most aud casity cuuccivcd

rsolution.

Thcn

aud

<),<. -shuwin~ wh.cii ~volves

with radius

thc~-hojc uni(ur.n)y, siu~

-stri.~I.sata~'tnonu.ntin :uni tliat cach pa.-Ucic Intact, t!.c its position wh.dc < ~tcni

one dc..scrib~ turns

j.Jano circi~ witi~ut t

<]isp)ace.ncnt. cachrevuludunin
quitc assise

rdativc

ahuut. H.ctimuT-

c.tui)ib.-iuni, c.mi{~tin<r Th~n~nic.s uf-t.hiscn.sci.s

rsultant p..rtiuu furce.

wh~t).cnmt.H)u!sc<~finu.) t.,o.K.phuK.,thc uf)L].c tensions at i!H.L.xt.-cn.iti~<,f arti.~ any s'n~t) oi' thc striu~ Latancud iu~Ht bL-ing by tlic cuntrifu.~]

as

144.]
144. 'ix. Thc gnerai

UNLIMITED

STRING.

179 for a uniform

cHScrendal

equation

stril)"-

cxprcssing

fho
A change

rciation

string. attcra.tion /or?~

in

thu

is propn.gatcd direction. positive a; and at thc tuue

form of thc a;andy, reprcscntsthe In tho value of t is mercly to an cquivaleut of x;, so t!)at (4) indicates that a certain origm witli uniform aiong thc string velocity ft in tho WImtcvcr thc value vainc thc point of maybcat of y will obt:uu at thc point so long as it

bctween

t, th samc a: + a A< at ti~e time + A<. Ttio form docs not Whcn
positive tion and thus

viotatc
titc direction,

perpctua.tcd thu rustrietioas

consists a certain

may ho any \vlu).tcvcr, on whici~ (1) dcpcnds.

of relation

motion

thc propagation
subsists betwccn

of a wave
thc

in thc
iuchna-

thc

velocity

at any

point.

Dif'ercntiatinn'

(4) wc find

Initia a.bovc by(4).

[y,

und

n:).y buth sati.sfied, t)ic

bc gLvcn motion

arbttrariiy, cannot

but

if tho

relation

Le not

bc rcprc.scntcd

Inasmultu'nmnucrthcuquatiun

y=~+~).(G),
dnotes the propagation butween of :), wave :Lnd in tho ?<e~(t<tM direction, to (5) is and t!)C relation

corresponding

]2_2

180
lu tho

TRANSVERSE

VIBRATIONS

0F

STRINGS.

[]4.t.

propagation
anu iL~

case tlie motion consists of th simuttaneous gnerai of two waves with vciocity ?, thc one in the positive,
in die uugative du-cc~L'u; aud tiiesu wa,ves arc

cthcr

entirely

indepcndcnt = Mt, =

of one a.nother. T)ie 10 Illltw, initial ioto two

lu

the first ~= ft~ <c' must and an

and ho ue

m t th e secon second conecived the relations will adva.nce second,

n~-

values va \lCS of

to be divided (5) and (7). in thc positive wave.

the negative

which satisfy rcspcctivcty parts, The nrst ccn.stitutcs the wavc which direction without the change ofform Thus, Initia)]y,

whence

If the the

disturbance

the string, interval cf time portion.

be origina)!y confined to a rmite of portion and wavcs positive t))0 ncgative sep:L:-atc after for each to traverse bulf the disturbod required

disturber). part initialiy side remains at rest nntil th positive wave has travelled ft-om A to P, is disturbed th passngo during of the wave, and ever after remains at rost. Th negative wave never affects P at ail. Similar statements apply, ?!H~M ~M~iA', to a point si de of~4Z?. If th character <3 on th negative of th original disturbance he such tha.t f<.c o a< vanishesinItiaUy. tho-~

for example, Suppose, A point P on the positive

that

is th

144.]
is and no if positive Cf~ +

AND NEGATIVE POSITIVE WAVES.

wfi.ve, vanisti the and th point there the P is never disturbed wave. are whei'e initially, and can Is no ngative C) waves cases If

18L
at all; K<

t (tM<

vanish and wavc

iiiitially, e(~un.l, and

positive ucither tlien

ngative In vanish.

similar eittier

its cvanesccnce to be due to the vanishes, may be considered of two one depending mutual destruction waves, ou componeiit on the initial and tlie other th Initi:d velocities. di.sptaccments, On thcy that th one side thse two wavcs Ttiis one anotlicr. destroy soonct' P can fail to bo affectcd conspire, th explains or later and on the other apparent paradox, aftcr -~jB Las been

disturbcd. Th
without

subsquent
vutocity, may

motion
he the

of a string
readHy ngative traced

that
by

is initially
graphical

displaced
mcthod.s.

Sinco
ncccssary

tllC

positive to dividc

aud t)iu

wavcs

and onc right, ()i.spi:(cc thse, to at, and then to recompuund a spacc equal them. through method to tho case of a plucked shall present)y apply this of nnite tongth. to ]-t5. Vibrations are called N~o?M)' when th motion

original to tho

disturbance

it is on)y equaL into two equal parts, th ot)ier to the left, We string

are

of each

to some functiou of th system is proportional. of th time, partidc If we endeavour to satisfy tlie same for a!l th particles.

Ly iissuming a. function

Avhore J~ dnotes y=~Y', of t ou [y, wc H)id 1 ~T Id~Y

a. function

of a? on!y, and

Y'=A'=~ sothat

~constant),

proving arbitrary

that

th

vib:).tio)is

must

bo

simple

pcriod.

Thc

value

ofy

mny

harmonie, be written

though

of

cos (~t~ y = e) cos (/a; a) PCOS (~(~ + M.T; e Ct) + ~7'' cos (Mf;< = that th most gcner:).l kind shcwing be regarded as due to the superposition

?~.K

e -t- a).(3), may vibra-

of stationary vibration ci' cqual progressive

~82

TRANSVERSE

VIBRATIONS

0F

STRINGS.

[145.

whoso tions, two stationary Th

directions vibrations

of propagation may eotnbinc

arc into

opposcd. Converscly, a progressive one.

solution in tlic ~=/(~)+~~+~ instance to an infinite but may bc interprcted string, so as to thc solution of thc prol)le.n give fur a nnitc in certain string ca~cs. Let us .suppose, for tl~t tl string cxamp!c, termintes at ~=0, a)id is he!d fast whilc it cxtcmLs thcrc, to inHnity in th positive direction Nuw so long as thc point on)y. .-c = 0 rcmains at rcst, it is a mattcr Mtua]!y of indinfcrcnco whcther the Le string on t!to prulongcd hidc or not. Wc ncg~Ive arc thus Icd to regard th gh'cn as string ibrnung part of one aud to scck ~iicthcr doubly inrin.tc, and how thc initial disp]acGmcnts a.nd vetocities on tlie ucgativc side can hc taken, so that on thc wltolo t!)crc shidi ho no ~=U dispfaccmcnt t!ic throughout motion. Titc initial subsquent values ofy and y on thc positive snic dtermine thc and ngative corrc.sponding parts of t!.c positive wa.vc.s, into which wc kuow that thc whulc mution can bc resolvcd. Thc former bas no influence at thc point .7-= 0. On th ngative S)de thc positive and th ngative waves are hntiaHy at our disposa!, but with thc lattowe arc not concerned. TI.c problem is to dtermine th positive wave ou th .side, so that in ngative with tlie conjunct.ion wave on givcn t).c positive ngative side of tlic origin, it sh:dt Icavu that point undisturbed. Lct
wavc m

M~

bc wluch

thc

line

advanees

(of any form) in ttie ngative

i-cprcscnting

th It is

diruetiou.

evident thc uthur

that side

th

reqmrcmcnts cf 0 what may

is tlie gcumctncid which ])a.s.s~ tijrough of O~

on by taidng be caUcd t!)c c?!<a?-~ wave, so that centre, biscctmg every chord (such as TV) it. Au:dytlc:d)y, If =/(.c) is thc quation is thu equatiou =/(-) of O~'Q'7);

of thc

case

arc

met

145.]

REELECTION

AT

FIXED

POINT.

183

Whon th

after right

a, timc

t the

curves

M'e shifted

to

the

loft

!md

to

a, distance co-ordinatca at, the throttgh rcspectivcly and opposite, and to ? = 0 arc necessa-nty cqual cut'rcspojiding tlicreforc when conipoutidcd displacomont. give zero rcsultant Thc efcct of the
t!):T,t a positivo th wnvc

coustrahit
ngative n.t tho

at 0 may
wavc s~mc tnoves timc

tttcrcforc through
merges

bo reprcsented. undisturbed, from (9. This pa.rcut by tbe

by supposing
but that l'cfL'ctcd iulluwing w~vu ru le

may

a,t auy

timc

be fouud

from

its

Lcit ~7rcflectcd wavc

bo tho is ttic

of the pM-cnt wave. position which this would assume, position

Thon

the

if it werc

turncd rotation,

two nght thro-ugh and then thrungh \vavc rciteetion

fn'st about OJC as an axis of angles, OY. In other thu samc angle about is thc iu itnagc O~Y aud of ~P()~~ OY, regarded formcd as by piano

t!)G rutuni words, successive optical mirrors. Th process. same lu rcsult

may

aiso solution

bc obtamcd

by

a more

analytical

the guneral

y=/(a;)+F(~+ft<), tito functious /'(~), J~(s) arc fur al! positive values ot' z.
dutcrmincd by

Th

condition

the initial cn'cumstances at = 0 requircs that

fur

al!

positive

values

/(-~)+(F(~)=0 or of /(-~)=-F(.)

fur

values of positive tc'nnincd ibr aH positive Thcrc ~o is now

z.

Thc

functions of and

and

.F are

thus

dc-

values

points of turbance in ~17~ dividcs which are l'cnuctcd

no difnculty in tracixg thc thc strmg -/i and J? are hctd itself Into backwards positi.vc and furwards

course fast. and

ofcvcnts Thc initial

wbcn dis-

wavcs, ngative bctwucii tlic nxed

184

TKANS VERSE

VIBRATIONS

0F

STUINGS.

['145.

their points, character from positive ch~nging to negative, and vice M)' at each renection. Aftcr an even numhf~ pf )~ tions in each case tite ibrm and original motion is compictely recovcrcd. The is most casiiy proccss followej in imagination when thc iniLiaI Ji.sturb.mcc is connncd to a .sma!l part of the more when its charactcr iitring, is suc!t as to particularly give rise to a wave propagatcd in ouc direction The ~~ travels with on!y. uniform velocity (f() to and fro along th Icngth of the .string, and after it has rcturned ? ~eco?~ time to its the starting point' condition of is exa.et]y rcstorcd. original things The period of th motion is thus tha time for the to traverse requircd pulse the length of the striug twicc, or

Thc of

s~unc the

iaw cvidcntty

ho]ds

good

wlmtcver

may

be the

character may part

happen of T. 14G.

original that th

disturbnncc, only in tlie s/io?'~ of rcurrence period

case it gcnera! is some aliquot

may bo advantageto thc case of a plucked ons!y applied Since the initial string. haf of tho displacemcnt vanishes, velocity to the positive belongs and haf to th negative wavo. The ma.nner in which th wave must be complotcd so as to produce the same effect as tlie conis shewn in th figure, straint, wliere th uppcr curve rcpresents

TItc metliod

of the

!a.5t fcw sections

tho

positive,

and In order

thc

lower the

the

negative

wave of the

in

their string

initial at any

positions.

to find

connguration

146.] fotm'f has space

GRAPHICAL :nust the two curves time, been sluftcd to th right and e(~ual to at.

L%IETI-IOD. be

185

after the upper superposed, tbe lower to the left through a.

TI)G resulta.nt pices.

is made curve, like its components, of a, twcifth A succession of six at intervals

up of stra.ight of th period,

shewiug taken
through

tho
from

course same be

of th

Helmholtz. thc stages'.

is given vibration, the string From

in the goes

figure (FIg. 27), back aga.In to il

It points 147.

will of

observed

that

th

inclination two be madc

of th constant at th

support

alternates small

bctween

string values. time

at

tho

If

disturbance

t at

the

tlic effect will not be fcit at stretched string, point x of an infinite will be in ail aftcr thc lapso of the timc 0 until a, and at disturbance had bccn made the same as if a like respects tliat similardisturbthe point a; + Ax at time t- A.c-r a. Suppose an ces
points

are
whosc

communicated
distances

to frorn

thc

string 0 incrcase

at

intervals time

of time by ctS~

at

each

then

1 This mothod. of troittUf; tho vibration of a plackod string is duo to Yonng. J~tt!. 2'Mf~ 1800. Tho studcnt is Tecommonded to mnkc Idmself fftmi!iar with it by actuaHy constructiug th forms of Fig. 27.

18

TRANSVERSE

VIBRATIONS

0F

STRINGS.

('147.

it is vident intorvais

that

thc

result

at 0 will

be the

sarne

as

if tlic

dis-

~nc~w<~anniauuatth.!san)upoint,j)rovtdu(t~iatti!Ctii:i~

bc incrca.sed fru.n T to T + 8r. This rcmark contaiu.s tlic of pitcit duc t!)tjoiy of t)to altoi-atiun to motion of tixj s(n)i-ce of a subjcct w)uc)t will corne uuder disturbance; our notice a~aiu in connecti'Jti with acrial vibrations. 148.
vibration, AViten trains onc of

point
wavcs

cf

an

innnitc

string

i.s

subjcct

to

forccd

bot!)'directions acoord.n~ to hws, wf.ic)i arc roadity Wc shall snjipo.sc invost.i~ated. tb~ thc or~in is thc point of excitation, th string thcro bcing to t!~ forccd motion subject and it will bu suniciunt y=~ to con.sidcr tl.c positive sido. If tbc motion of cach dc.ncnt bo resistcd force by t)ic frictional tlie dinTercutial is quation

procccd

frorn

it

ill

148.]

WAVES. DAMPIN& 0F PROGRESSIVE

that /<:is small,

187

If wc suppose

:uid
sttcws

This

snhttion

thft-t thcrc

'\v))ost; funptitmtu stowty t)iis If <=(), cxponcntiaHactur. a wavc, simpty

ia propa~tcd the string a!oi)g duninishus oti nccunut of ttie factor disappca.rs, iuid wc hve

This whcn
to a

rcsult

stands

in

contradiction thc forccd

must

to

t)ie

thcre
sing)e

is no friction,
snnph harjnonic

vibrations
bc contrM'y,

!aw gnral uf a. System

in

H)a.t, (due
ptiase

force)

synchrouous

varies phase to a.uother along tiie string. cuntiuuuusty to suppose Thu fuct is, tl)a.t wc M'o not a.t liburty /e==0 in (8), on tuu assumption that was obtalucd as timt cqmt.tion Inasmucti aud not zro. Howcvcr thc rca,l part of X in (3) is positive, long oi' friction t!m coenicicut a nulte 'bc, may Le ta~en so stnng may tbrougbont. According in passin~ bufurc t!t0 dampcd readiing reficctcd waves bcgin. tliis poitit of smaUness~ Hirthcr end. Ai'Ltjr is dinilnistied and when thc friction to compHca.tc th result, inust be takeu sries of such iuto accoun.t, an iunnite indonuitely, of thc samc phase motion and wcuid give a. rsultant throughout. stnaM that the vibrations are This to be Thc it will problem concuntratcd bc found may at that be soived for a. string whose mass is supposed by th mcthod of 120. points, C(p)idistant to hc givcn (= ~le""), and may be supposc<) thc systcm of quations (5) of 120 may a.11 not

to (9), on tho ft'om une point

t)iu

co-oi'din:Ltc

be satisned

by taking

whcrc tn)!i.

is a comptex Thc result for

constant

dctemuucd string

a. cuntmuous

by a quadratic cquade)nn-y bc afterwards

duecd.

CIIAPTER

VII.

LONGITUDINAL AND TORSIONAL VIBRATIONS 0F BARS.

in order of simnjicity string is usuaUy by undcrstood in Acoustics of natter of uniform substance and c)ongatud mass cytindrical orm. At t!ic c..ds thu is eut oH' by p]anes cylinder pcrpcndicuJar to tlie lincs. Tho centres gc~-atu~ of u.c.-tia of t).e tmnsvcrse sections lie ou a stra.ght ]inc whic!. is calk-d t!.e
Thc vibrations <-ta bar

1-tJ. is tlie bar,

Tin,

next

sysL.m winch terni

to

th

arc the most thc

of

throe in

kind.s-iongitndina! ~-rorJ: tt.cory. Tbcy and will on]y and are bc

but ~T~' at considered

t.c

same

time

the
m

diHicu)t next

by thc.n.sctvcs rcferrcd to hcrcsofarasis W)th thc othcr two ktnd.s

chapter,

ncce.ssaryfor of vibrations.

co~parison

contrast

votions Long.tndu.at arc those in which thc axis romains whde t)~ nnmoved. transversc sections vibrato to and fro in the direction to their pcrpendieuL-u. Thc moving planes. is powcr tho r~stancc o~red thc rod to by extension or compression. class of vibrations OucpccuH~ityofthIs I.s at once vident Since the force to produce a given neccssary extension in a bar is proportional to tho area of the ~hHe section. th ,na.ss to be moved a!so in the same it fo)!ows t!mt for a bar proportion, of and given length tunes and the modes inatcrial ti.epcriodic of vibration arc ~dpendent of th area and of tlie for.n of th traverse as we shaU sect.on. A .sinufar law obtain.s, prcsentty in tite -sce, case ot torsionat vibrations.

Itisothcrwiscwhen the vibrations arc latral. Thc pcriodic tunes are mdecd of t!.e thickness i.~ependent of tbc bar in th direction to ~o plane perpendicular but the motive power ofuexurc.

14!).] in this

CLASSIFICATION

OF

VIBRATIONS.

189

to bcnding, more rapidiy incrcases cttse, viz. tlie rsistance th~i the in that in thickness and therefore an incr~lae plane, tinckuess ]s accompa.uicd by n risc of pitch.
mcchancase of Iongi.tudiun.1 ttic and latral vibradons, ical consta.uts coticcrncd a.rc thc dcnsit.y nad tho of thc m~terud v:due ofYoung's For sm:d) extensions tnodulus. (or compressions)

In thc

Hookc's hoids bc

Ia.w, according good. Tf ,i If tLe

to w!dch extension,

thc viz.

tension aetnid

v:n'ics

a.s

th extension,

c, we liave y=~, is th tension m'en, per unit moJnhis Young's maythercforc to a bar ha-ve to bc appHcd length, sions; area. The stant place. if Hooke's law its dimensions

callecl

whet'o
ncccssary

nntural ]ongth -n ",i 1 nittundtengtn and T is Young's nioduius, Icngth

e. to producc thc extension be dcancj as tlie force whieh would its of unit section, in oi'dcr to doub]c to hold those good for so grea.t of a force divided extenby an

contiuncd

are a.ccol'd.ing~y

torsional

vibrations interprtation

IL, whose

depend will

aiso be

on

a second in

clastic the

conproper

considered

classes of vibrations, tlleory the threc depcnding on rsistance to extension, to torsion, and to ncxurc respectively and independent of one another so long as th are quitc distinct, of the strains may be neglectcd, squares yct in actual exprimenta Although with bars which are ncititer
uuiform in matcria).

in

in figure it is often found cyliudrical tudinal or torsional vibrations withont measure of latral motion. In bars

Impossible tlie accompaniment

accuratcly to excite longiof some tb

nor

lateral motion is far graver gravest and that or torsional it will generally motion, consequently happcn of th latter kinds agres more or less thc principal tonc of either in pitch witli some overtone of th former kind. Under perfectly such circumstances and a small thc rcgidar modes of vibrations effect, in a bar becomc Thc is uustabic, irregularity a great may prcduce vibrations longitudinal to vibrato proceed
no

of ordinnry dimensions than tbo gravest. longitudinal

of exciting dimculty purely similar to that of getting a string With classes this

in one plane. to consider tbe threc

we may explanation of vibrations independently, which those will in fact

raise

cominencingwith
mathematical

longitudinal questions chapters.

vibrations, beyond

aiready

clisposecl

of in th

previous

190

LONGITUDINAL

VIBRATIONS

0F

BARS.

D.50.

a rod is stretchcd to its tcngth, by a force parallel the stretching is in gnral accompanied contraction in by latral su~-h a manner tha. th ~<) of v<Juiuo I~ss than if th dplacement of cvcry particlc wcrc paraHc! to thc axis. In the case of a short rod andof a partiel situated ncar tlic cyliudrical this L~tcral motion would bu bonndary, iu comp~ble jn~nitudc with thu longitudinat and eou]d not bc ovorlookcd motion, without risk of considrable crror. But where a rod, whosc is grc~t Io)gth in proportion to tho lincM- dimensions of is subjcct to its.section, a strctching of onc sigu t)ic longitudinal tliroughont, motion accuand multes, thus In th caso of rods ordinary vibratmg lon.din thc graver thc tu()in:d!y modes, inertia of thc hter~motion may bc negicctcd. Morcover wo shall sco lator how a correction if necessa.ry. may bc introducud, Lct section bc th from distance of th
hyer of particles composino-

150.

Whcn

unstrctchcd, vibrations, position of a,

thc cqnHibrium position cithcr by pcrtnancnt an<I !ct by bc + laycr is thc section, t]ie
T)~

of onc tension

end, whc.i or as thc so

and

any thc~-od is rusuit of actuat

position

<Hsp]accment,
eqnihbrium

that
actuai

th

is givcn nuighbounng tl.c e~~

bcing und

a;+~ thus,

~-+~+~+~~ if T be thc

f~ tension

rc-

spoctivcly, unit

per

arca, acting

across

Considcr and by tho + 8~. (1) y~ tension is

now If tl.o

the arca

forces

on acting of tho suetion ngative

thc bu

s)ice the

boundcd tension and at

by a: at .c is a:+~

actiug

In thc

direction,

~~+~
in thc positive acting due to th action of direction; th adjoining and parts

J thus on the is on Die whutc th force sHcc

Tho and

mass

ofthc if ~be

clment

is

thcreforc

titc acce!erating

If p bc tl.e original density force acting on it.thc equa-~

150.]
tion

GENERAL DIFFERENTIAL EQUATION.

is

191

of oquilibrium

In of:ui

what itnprcsscd

foHows
force.

wc shaH

not

To find

thc to cojtsidcr rcquirc of motion th quation ~gaiust acclration

opration wc hve aud

ouly thus

to replace if p =a~,

by th wo hve

raction

Tins transverse undistnrbud

ngative condition

quation
displacuments

is of th
of

same

form

as

tbat

a, strctched
waves of any

string,
type in

to applicable an<) itidicatcs ttie

tho thc

propagation
directions. of

of Ttie thc

velocity apparent spacc wilt

<t is rotative

and positive to thc UH~'e<c/<er/

thc

bar;

n.uce

is

strctched distinction 151. wc liavc

lu propugatud aud uustrctched is matcrial For tho

vuloeity witb which a disturbin th ratio of th bc gruatcr of any portion

case of permanent

Icngths oniy in t!ic magnitude f~ = </

of thc

bar.

Tho

tension.

actual

of thc p = ~M tension

vclocity ~)M,

of propagation,

to necessary (according of th bar and t)t0 longitudinal Hoo~c's law) to double thc length wit)t total tension bar wcrc strctchfd If tho samc T, density. of wavcs alung it of propagation th velocity and wcrG ncxibic, that thc vclocity bo titcn would hc ~/( 2' /3M). In order inight Le ~M, or, in othcr Tmust words, tlic thc 8:nnu in titc two cases, which is the ratio of the wholo tension double arc thus would tho hve to in bc t)iat toncs Icngth. very high Thc
nccessary thcorcticaUy of longitudinaUy

in ordcr vibrating obtainable

to rods

comparison

with

tilose

froin

strings

ofcompiu'abtc case
s<p)are

Icngttt. thc viduc

In titc pcr weight of force on thc same

of stcel

thc

gravity

syston rcferrcd

of q is about 22 x 10" grammes utnt.s To express this In absotnte centimtre. In wc ninst c. f!. S.' systmn, mnttipty by 9SU. witb its spcifie thc dcnsit.y of stcct (Identical fur steel to water) is 7'8. J~-nec

1 Centimtre,

Gramme,

Second. Brit. Ase.

Tliis

of tboDritiBliAasociatiu.

System is recommeuded 1873. Report,

by n. Coinnutteo

102

LONGITUDINAL

VIBRATIONS

OJ

BARS.

fiS].

~ut.

0,000

cent.n.etre.sper

velocity of or about second, 1G ti~es

-~el grc.ter

is

Lhc L samc It

"T in .stec]. as to bc nicntioncd

expc.ri.ncnta th c~c of ga.cs,

ought

th<it in strictnoss
is not that

minedby.statua! hre As in chaptcr, tion ofsounJ

to

t.hc value wiuch o~ht

dctcrnf tu be ~scd

.vbi.~ witli

will

bc. trcntod co.~crnc-d e~cts,

thc

mp,d altcration.s arc tende.! n~e thc active on c.xtcn~.

arc not

of state

lu a .subscq~nt iu t)~ pr.p~onc rusul~of

~nch from cbscrvat.~s turc But tho d.ta ci any consGqucncc

ther,~ va!ne cf

co.luct.cd
prcise enoug]~

ni tlie

tl.at obtaluod bcyond at a constant tc,np~ to m..d<c this con-cction

c~sc of solids. '<~tud;na] bar, n:unc]y

v~ vibrations

ofau

uniimited

bcing eonsidered Whcn nent

the

same hcre. both

~=7(~-a<)+~(~+~), as t~t to a string, appHc~Ie ends of a bar are fre~ titere at the ends the.n.sdvc.s

need

not

be furti~r no pcrmitcn~rv

and tcns.on, tension. Thoconditiouforaf.-cec.ndisthcrefore

is of course titerc is no

~=0.
le doter~nc vanes th~t~ as a function t)~ nor.ual as a harmonie of .r, ,nust n,odc.s function satisfy of vibration, wo must of tho timc-cos7i~ assume TI~n

Now sinco~vanishcs
again smce

al~ys ~~cn tih.Lt

~hcn ~=/-thc

~=0,

we get j3=0; iungth of

an<! tlic

g sin ~~=0,

vanishcs wbich

u.turat form

bar,

sl~cws

is oftiie

t'bcingiutc'graj.

152.]
Accordingty, form the

BOTH
normal

EXTREMITIES
modes arc

FREE.
given by quations

193
of thc

in

which

of

course

au

urbitrury

constant

may

bc

nddcd

to

< if

<!esh'cd. Thc complete by solution for Il bar with both ends frce is thcro-

forc cxprcssed

whcrc in the givcu.

arc arbitrary and usu:d whun mauncr,

constants, thc Iuiti:d

wliieh

may

bc

detcrmincd

values

of

aud

arc

A zcro

vainc

dispIn.ecmGnt aud amounting TIic

of i is admissible it gives a term a. rcprcsentmg constant with both to spn.cc and tuno, respect in fact only to an altration of the origin.

of the gravcst in (6) period component to corresponding is which is thc tinlc t=I, 2~ in occuhied by a. disturhanee twice the Icngth of t)io rod. The other travelling toncs fonnd values to i form a complte Ly ascribing harmonie integral scale i so that to tliis tl)c note according theory givcn hy a rod in longitudinal In thc place
pression

vibration

would

bc in aU centre but thc

ca~cs

muslca.1.

of

mode thc gravest no motion, or nodc; f~

is thcrc a maximum.

of the periodic

rod,

whcro elon~ation

/c= or

is a com-

153. may ends initial be

The deduecd

case ofa from

bar th

with

onc end For n,nd frce

frec

and for at both

the

other with

fixed both

gcncral

solution ftxcd a.t

a bar

the Icngth. froc, and of twice statc of thc bar froc ut .B=0 a.ud velocitius from of th ma.y a.!ways 0 to 2~ and

whatever

placements

bo ascribed

!rmy be ttie x = l, sucii disto the sections ends as shaH

from 0 to Identical in th two parts cases. It is only ncccssary to suppose that from to 2~ the disand vclocitics arc initially to thosc placements cqual and opposite found in thc portion from 0 to at an cqnal distance from thc ccutre
tl)c

ofabarextending make thc motions

x =

Uiidcr
rcm!).in

thcsc at
rest

circumstanccs
throughout

tho

centre and

must thcn 13

symmr-try

t)ie motion,

by thc

R.

194

LONGITUDINAL

VIBRATIONS

0F

BARS.

H 5 3.

portion from 0 to s~tisnes all tl.e required conditions. We conc udc that the vibrations of a bar frce at one end and fixed at tho othcr arc ~dcnfj.d ~L .h. of bar uf twice the longth of which both ends arc free, thc latter vibrating only n unevcn modes, obtained in by ~king succession a))~ i~~ T!~ tones of tlie bar still te a hannomc belong ~.h, b~' cvcn toncs (octave, &c. of th are .vauting. fund.menta!) Th period of the tone is th time gravest occupicd by a pulse in travelling/b~ timcs the length of th bar. 154. be satisfied At K-0, constant dep.n.).nt)y we should at once. method. that Assuming as a fonction ~COS7!f~ we sec that as a function ofthe .h ~)cn both ends of that we may suppose is zero if that j a bar the are value ~=0 the

fixed, ~t~ At

con(litions is tension.

to

&

a small 1 "'1:"

no permanent of' th. vibrations w.),av. ~idcnt)y obtain our result most simply bt~M' But it may

f~T< this tenu

timc

varies

as

+ 7?sinK~, satisfy

of x it must

of which

the gnral

solution

is

J But since we may write vanishcs with x for ~11 values of t, <. r C'=0,n nnd ~) thus

154.]

BOTH

EXTREMITIES

FIXED.

1!)5

The
which

series ])owcvcr case of

of

tones

form
of the

a,
mcmbo-s

compote
may

harmonie bo mi.ssiug

scalc ni

(ft-on any

any

of the gravest comperiod is the poncnt by a pulse to travc! twice tho Icnn-t.b of th rod, thc sa)nc thcroforo as if both ends wcre frec. It nnist be observed thnt we hve bore to do with thc MH~r~c~~ of th rod, and that th length period for a givcn natural length is ludependent of the permanent tension. vibration), tinic takcn
The case of solution no permanent of

actua!

and

tho

the

problcm
tension

of
might

the aiso

doubly
bc

fixed
derived

bar
from

in to

the
t)iat

of a doubly For in thc

free latter

bar

by mcrc

ditfcrfmtiation satisfics the

with necessary

respect

.c.

problem

difereutial

quation,

viz. (I'

~.E~
d

= a2 (le <

masmuch

ns

satis~cs

vanishcs. According!y (lx satisfies ail tl~e conditions prcscribed for both ends arc ~xcd. The two sries of toncs 155. the rod Thc effect of a small ioad

and

at

both

ends

dx in are

iu the thus

this caso

problem when

identicul.

~f attac])cd

is rcadi]y ca!cu)ated as approxnnatc!y, to assume thc to bc uuaitcrcd type of vibration will takc the case of a rod nxed at .~=0, and free kinetic is proportinnal to cncrgy

to any point of it is sufncient ( 88). at .t'= \Vc The

or to

~G Since ciples
'h~ucc compouent

LONGITUDINAL the potentia.1

VIBRATIONS

OFDARS.

[155.

is uudtcred, we sec by t!ic prmcncrgy of Chapter iv., th~t tho cfrcct of ti~) sma!) !o~ at a &: u'u:ti Lhe iixcd cud is to inci-cMC the period of' tho toucs in thc mtiu

Tho wholo

snrnH

mass

quantity of th rod. bcatt~chcd at

p~

is thc

ratio

of

tlic

!o:td to t!ic

Iftheload effect

thc

frec

end,

sm'~=l,

and

tlic

is to dcprcss the piteli mtcn'd. It will bc rcmembei-ud If the tlie pitch 150. thc In frcc
th)3

of cvery tone by t!ic s~mc small t!)at i is hei-c an MMC~t mtcgcr. of J!f bo nodc uualtcrcd notice
with

point of that Another end is

wo

of .~chmcut eojupouont problem grc~t

inay iu

ronams worth
compiu-isoi

of any componcnt, by th addition. whcn masa t]to of load thc ~t rod.

occurs thc

CMC

assume along

of Utntbrm If

extension

type of vibration, tlic Icngth of the rod. of thc load 3/, tho kinetic

as thc

a. condition

bc tttc

displaccmcnt

cncrgy

is

correction duc to the to t)ic dition to ~ofone-third 1.~7. nmy tttc latc.ral close Our mathem~tic~ with an estimatc
n.ut.ou of

The

incrtia

of the rod is thus cquivalent of thc mass of the rod.

of JongitudinfLl

discussion

vibrations in ncgieetinrr situatcd on tlic

of thc parts of

cn-or th

invo!vcd rod net

thc

157.]

CORRECTION

FOR

LATERAL

MOTION.

197

~xis.

If denutcd

the ratio

of latral

contraction

i,inj huerai bj. ?- from thc axis will bc ~re, in the cxtcMsion. in strictiless this relation Altiiougli will bc modiiicd by tho Incrtia of thc httcl-al motion, yct for thc prsent purto hold good. pose it may bc supposed TIic constant If/~worc uumerical /t is a longitudinal ucgativc, and if were greater to quantity, tension tbau 0 and lying between would produce a latral tlie lateral contraction th elongatiou, and cause latter

to longitudinni cxtoisinn of a. particie disp~acoment distant the case of cquilibnum, where e is

statc of thin~ aud tlie former can stability, scarccy be possible in ordinary solids. At one time it was supposed was nccessarily to that so tliat equal thcro was only one clastic but oxperimcuts independoiit constant, have since shcwu. is variable. For that glass and brasa Wcrthchn found expcri= /t nicutally witli If from the dnote tlie lateral axis, and if thc duc to t]ie lateral motion is displacement section bo of thc circular, distant r particlc th kinetic encrgy

swelting, would bc grcat oicugh a diminution of volume would be mconsistclit

ovcrbalance on the whole.

Th

Thc Inercasc

effect the

of tlie

incrtia

of th

latcra!

motion

is thct'cfoi-c

<o

poriod

m thc ratio

This bars

correction of oi-dinary

will

bc nearly insensible for tlie of length to thickness. proportions

graver

modes

of

198

LONGITUDINAL

VIBRATIONS

0F

BARS.

[158.

158.
v.'itii icds

Exprimenta
of dc~l r

on longitudinal
ot' g!a.~s. Tho

vibrations
vibmtbns

may
arc

be
cxcitcd

made
by

friction, wooden rosin. cxcited

with rods "T!tc

a wet

doth

it is neccssa.ry longitudinal

of glass; but for mtal or to use Icather with powdered charged vibrations of a pianofortu string may bc

in

the

case

it longitudinaUy wIHi a piece of india rubbing of a violin string the bow obliqucJy rubber, by p!a.cing across the string, and moving it aiong thc string loogitndina.Hy, tire same point of thc bow unon th strirtg. Thc note is kecping sin'ill in bot!i ca.sus." unpjcasn.ntiy by gcnt)y and those so as to attc't' thc pitcb of peg of thc vioini bc turncd th lateral vibrations it will be found tba.t t))u vcry considcrabty, vibrations )ias :dt(ired vcry shghtty. Tim pitch uf' th ]ongitudina.i uf this is tha.t in thc case of t)tc lateral rca~on vibrations thc of of wavc-transmission on t)io0 ehtuigc vclocity dpends cbicny change of tension,
which is considrable.

"If

t]te

But

in

thc

case

of thc

longitudinal sion dpends sligttt'

vibrations, upon th

of vclocity of wavc-transniischange which is comparativcfy change of extension,

thc

on longitudinal a, peculia.r expci'Imcnts vibrations, calted a "son rauque," was occasionaMyobservcd, sound, hyhim whosc pitcii was an octave below tl)at of tbc longitudinal vibration. to Terquem" thc cause of this sound is a transAccording verse whuse is due vibration, to an appcarance approximatc agrecmcntbetwee)i longitudinal wuld be that Itsown vibration. one of th of thc pcriod and that of the sub-octave If this view be correct, the phenomenon second referable to the fact order, prubabiy of a bar tends to produce curvature. whic!i small

In

Savart's

longitudinal 15!). Thc on t!te

compression second class

dpend

rsistance

of vibrations, ca)ted to twisting, opposed

torsional, is of very

A solid or hoi)ow eylindricat rod of circular importance. section at the cuds, in suctt a may be twistcd by suitable forces, applied nianuer that cach transverse section remains in its own plane. But if thc section be not circular, th cneet of a twist is of a ]norc compticated the twist being attendcd cliaractcr, necessarUy of th of matter by a warping tho layers originally composing nornud sections. tho enccts of th warping Altijough might proDoukin'H ~'f. ~c')t~t'M, C'Anott-, p. Lvn. ~i. 12U1!)U.

159.]
bably
'.vhi!c,

TORSIONAL

VIBRATIONS.

199

be
v~jh:I

dctermiucd
c~nnti.

in

any
our~ivc~

particular
iicrc t,o

case
dm

if
c~e

it

wero
ut'

worth
c~-cutai

section, Th clitstic note

wheu force constant

there with

is uo motion which

pa.r:~tel

to the

axis

ofthe

rod.

different

it by n, tlie relation

is resisted twisting depends upon an from q, ca.Hed th If we derigidity. between q, m, a.nd may be written

shewing M=~.

that

n lies

betwcen

and

In

the

case

of ~=~

us now suppose that we hve to do with a. rod in the form of a thin tube of ra.dius r a.ud thickness ~r, and Ict dnote t]io of any section, angular displacement distant a: from the origin. Thc rate of twist at a: is reprcsentcd the and pipe since by by The area and th shear of the

Lot

material umtof round

composing area the is~ axis is

r~. thc

opposing the

force

per

is 27n-~

moment

Since cyliader The

is independent of fmitc thickness

this

of r, the same equation or to one solld throughout. la A/ vibrations, and

appUca

to

velocity similar

of wa,ve propagiition to that

the wholo the applies

thoory

is prccisely

of longitudmal

condition to inotropic

TIiornson aod Tait. 683. mntcria! on]y.

This, it ahould bo remarkcJ.

200
for a free

LONGITUDINAL

VIBRATIONS

0F

BARS.

7/]
end twist bcing 'C = 0, and for a fixed = constant. end be coutempla.tcd,

[159.
= 0, or, if a

permanent The

of longitudinal vibrations is to tliat vclocity of torsional vibrations in tlie ratio or ~/(3 + 2~) I. Th samc ratio to the frcqucncics of vibration applics for bars of cqna! Icngth in vibra.ting modes under corrcsponding corresponding termint conditions. the ratio of frequencies If == would bc :=~/8 correspond ing to an interval tbc V2
Longitudinal

:3=1'G3, ratitcr grcatcr must than a nftb.

In any case

ratio

of frcqucncics aud ~/3 vibrations

lie between

1 = 1-414, torsional

1 = 1-732. were nrstinvcsti"-atcd bv

and

Cbladni.

CHAPTER

VIII.

LATERAL

VIBRATIONS

0F

BARS.

160. vibrations straight. haps tlie Thcre portant

IN

wc sliall consider the lateral present chapter condition arc of thin ctastic rods, which in thcir natural is perNext to those of strings, this class of vibrations amenable to thcoretical and exprimental into prommenco theory, which treatment. somc imth fami-

tho

most

sufncieut to bring is dimculty connected with th gnerai points

with circular functions of th reader may lead him to pass liarity to strings; while at the same over too Hghtiy in th application arc not such as to engross attention time the difficulties ofanalysis to general matliematical and physical which be devoted should principles. Daniel th problem. Bernoulli' Euler, scems to have been tlio first who more attae~ed

Strehiko", who have 161. the Thc

Lissajous", advanced our knowledge Thc

Poisson, Riccati, a.nd A. Scebeck~

a.nd Cauchy, arc foremost

reccntly among thoso

of it. into two to according tension. longitudinal parts, additional complito stretchcd alneglecte to the tension,

divides problem or absence, of presence, of permanent considration and whose is of interest only stiffiiess, attention

itsolf

a permanent tension entails in its

cation, strings,

Our together. two extrme cases, (2) when the tension

small, though will therefore bc given principally there is no permanent (1) whcn is thc chief agent in the vibration.

application bc cannot

/<M< J'<'<r~). t. xnt. C'oHUMn<.

~;)~. f!. Chimie (H), xxx. !}85.

rogg. ~;t)!. Bd. xxvu.

~h/!))~~f~<'M d. ~/<!< sc/t<fc)t. Leipzig, 1852.

J'/ty~. Classe fL /C..S'<M'/. CMC~M/t~/t d. !rfj!C;t-

202
WIth

LATERAL

VIBRATIONS

0F

BARS.

[161.

to thc section of thc rod, wc sha)! respect that suppose one principal axis lies m thc phuic of vibration, so that t!ic bendinoat cvcry part takcs p!acc iu a direction of maximum or miuimm~ rtcxunU ri~idity. (01 st,atinary) For uxample, thc surface of thc rod may bc onc of rvolution, cach section bdug circular, thoun-h not ncccssarily of constant radius. Under t!icsc circumstances thc of th bending potentiaJ for each clment cncrgy of lungth is proto the square of thc curvaturc portional multiplied by a qnantity on thc matcriid of t))e rod, and on thc moment dcpcnding of inertia of thc transvcrsc section about an axis its centre of t))rough inertia to tlie plane of bending. pcrpeudicuhuJf be thc area ofthe its tnomcut section, of inertia,~ Young's th moduius,~ clonent of icugth, and ~F' t)ic corrcspouding poteutial for energy a curvature 1 of tlie axis of the rod,

is readily obtained the extension by coDsidermg of th varions filaments of whicli the bar may Le to bo supposed made up. be tlie distance from the axis of thc Lot projection on th piano of of a nl&ment of section bending ~M. TIien thc of the niament is altered length in th ratio by the bending

This

resuit

-K being th radius of curvature. Thus on th side of thc axis for is positive, viz. on th o~c~ whidi~ side, a filament is extended, while on thc other side of th axis there is compression. Tho force necessary to produce th extension by the deiiiis (~ tion of Young's modulus; and thus th whole couple by which th is resisted amounts to bending

if &) bc th area of th section and < its radius of about gyration a Imc through tlie axis, and to the plane of perpendicular bending. The angle of bcuding to a corresponding length of axis ds is aud thus the work rcquired to bend o~ to curvature 1 Ti! !q c ~t siucc th Mea?; is hdfthc~~ value of tlie couple.

161.]
For Th~t th

POTENTIAL

ENERGY

OF

BENDINQ.

203

a circular

section

? is onc-ha.If

t!te

radius. bc proportionn.1, is vident bcfore-

potential Ct'icrgyof thcbcndingwoutd to the squ:).ro of thc cut-Yidure, cc~e?'tN ~ft~tfs, hand. If wc en!! tho couificiotit J9, wo may tako

of tliat point on thc axis of y is tlie lateral dispiffcmcnt thc rod w!iosc abseissa, mc'asurcd to th undisturbed paraltel posiIn th case of a rod whose arc similar and sections tion, is x. 7~ is a constaiit, and may bc removed from under siniiladysituated in which the intgral Tho sign. derived lments elements partly from

rod is cncrgy of thc moving of translation, tlie motion to of th parallel from tlie rotation of th same it, and partiy thcir centres of inertia through perpendicular tion. Th former part is expressed by

kinetic

composing axes about of vibra-

to th plane

if p dnote the volume-Jensity. that th angula.]' only to observe ",a.nd (~uantity clement, therefore must tliat its angu!a.r

To express displacement vclocity

the

latter of thc Th

part, we hve dx is lment square of this of tho

bc multiplied is, by ~m

by haf t!ie moment < We thus obtain

of inertia.

1G2. ourselvcs confine

In

ordcr

of tlie ourscivcs

of motion we may avail equation of virtual we If for simplicity velocities. principle to the case of uniform we have section,

to form

thc

204
where th terms

LATERAL

VIBRATIONS

0F

BARS.

[162.

from the Intgrt sign are to be t~cn hetwu th I.nuts. This expression inctu.ics ouiy t])e internat forces due to tlie ben<hng. In what futlowa ~ sh.U! .s..pposc ti.at there are no forces Mting from wltlinut, or ra-thcr none that <)o work upon t))c System. A force of suc)i con.stramt, as th!it to ]iotd ncccssary any pu.nt of the hn.r at rc.st, need not bc it do~ no rcgn.rded, work and therctore cannot appcar in t!~c quation of virtual veioettics. Thc virtual moment of tlie acclrations is

free

Thus

tlie

variational

quation

of motion

is

in which between

th th

tcrms limits.

free From

from this

the

mte~-al we Jer-

sign

arc

to h.. takcn

~edatallpo~ofthel~t~f~~

Jongi ..J~.

tud

inl11

lVilVl'S.

1G2.] ]
Thc forma conceive condition according

TERMINAL

CONDITIONS.
at thc ends case. thc ratio assumes

205
different to

(5) to be satisfied to thc circumstances of such a nature

of tlic that

It is possible 8 ( ")

a constraint

8v has

a prescribed is then finitc value. Th second condition boundary obtained from (5) by introduction of this ratio. But in aH the cases that we shaH hve to consider, there is either no constraint or th thon th constraint boundary is such conditions that eithcr take 8 or [-") the form Sy vanishes, and

We an end becomc

must

now distinguish ( ~)

the

special

cases

that

may

arise.

If

be frcc, 8y and S

are both

arbitrary,

and the eonditiona

the acts

first

of which

at thc frec If th

may bc regarded end, and tlie second at th end

as expressing that no force

that acts.

no couple

be frec, but the end itself he conto romain at rcst by the action of an applied force of the strained word the in which case for want of a botter magnitude, necessary th conditions are rod is said to be supported, direction

by which A third

(5) is satisfied. case crises w!)cu to mainmagnitude,

is constramcJ an cxtrcmlty tttin its direction couple of the necessary by a.n applied but is free to take any position. We ha-vc thcn

Fourth)y, and position Thc

bc constrained extrcnuty may in which case thc rod is said direction, arc plainly conditions thc

both

as

to

to be c~n~ec~.

206
Of tlie Even romain clamped so groat third by with for means these

LATERAL

VIBRATIONS

OF

DARS.

[162.
important

four which tins

cases omit the

th to

first

and

last

are

the more are

we shall

consider,

as there

constraint contomplated a considrable simplification as but have cither the hccn end of complication expected.

no exprimentt could bc rcaHzed.

discussion,
as

or supportcd,
might

of problems varicty thc bar may bc frco, thencc arising is not We shaH thc from find that that of solution

difforent cases may be treated and that togethcr, for onc case may sometimes bc dcrivcd immediately another. In for a massive cxperimcnting end clamped construction. on may In thc bu vibrations rcahzcd with of

condition bars, thc thc aid of a vice of end thero is of course

thc case of a frec

no difilculty so far as thc end itself is concerned but, whcn both ends are free, a question arises as to how th weight of the bar is to be supportcd. In order to Interfre with the vibration as little as possible, th supports must be connned to th ncighbom'hood of th nodal It is sometimcs surHcicnt points. mcrely to !ay th bar on bridges, or to pass a loop of string round the bar and draw it tight to its ends. For more exact by screws attached it wou!d perliaps bc prcferabJc to carry th weight of purposes th bar on a pin travcrsing a holc driHed through th middie of th thickness in thc plane of vibration. Whcn contact longth 1G3. position, intcrfcring th terms an with end is to ba a fixed 'supported,' whoso plane it may be pressed to into the is perpendicular

plate

of the bar. Before which with procccding will greatly th value fnrthcr we shall introducc without sliall of the to a sup-

thc analysis, simplify of tlie solution. We motion amounts nt its

set-iolisly assume that sections of the shall init is

on th depending anguhu' th bar be neglected, which may tHer~ of' each section conccntratcd afterwards ertia, mail.

supposing centre. We for rotatory circumstances th

a correction ( 180) investigate and shall provo that under ordinary Tho quation of motion now becomcs

')M.

163.]

HARMONIC

VIBRATIONS.

207

Th thc take

next

assumption

step in conformity of the harmonie

with form

th ofy.

will be gnera,! plan We may convenicntly

where whose obtain

l is the value

Icngth ha.s to

of be

ba.r, and deternimed.

th

w is a.n abstract Substituting

number, ni (1), wo

p"* If M == e be roots solution of is unity,

a, solution, we see viz. +1, 1, +t,

that t;

p Is one so that

of tlie the

fourth

complte

containing

four

arbitrary still to

constants. thc satisfy detcrmino four th boundary ratios A conditions,-two (7 -D, and

Wc have for each end. furnish besides

These

an equation whieh '?~ must satisfy. Thus a series of particular values of w a.rc alone a.nd for cach ?~ admissible, th coiTcsponding ic is determincd in everything a constant except Wc shall the different u befunctions multiplier. distinguish longiug Thc the to
the sa.mc systcm

by suffixes. any time If may < bc cxpanded in a series of co&c. be tho normal

value

of y at

functions we have

( 92,

03).

ordiDates,

and

n.t this stage that each asserting functions and tl]crcforc th vanishes, product intcgrated of th followiiic, section need as more not bc regarded process than a Mr(/?ce[<t'o?t. It is however in order to determine rcquircd th value ofthc intcgra.ted squares. fully justified of the

We

arc

in

308
IC't. respOQding Lot

LATERAL

VIBRATIONS

0F

BARS.

[164.
functious cor-

M,

<)cnotc

two

of thc Thcn

nonna.1

respectivciy

to ?~ and ?/

If

wc

snbtmct

rcspcctivcly, we!ia.vc

tl)cm hy (2) aftcr multiplyiog cqn:ttiot)S over thc lungHi uf thc and thcn ixtugmtc

M~ biu',

the Intcgratod Now

tcrms thc

bcing end

takcn in

bctwecn bo

tlie

limits.

or cla.mpcd, supportctt, eacli terni vanishcs of one or other free', of its factors. We may therefore couchtdc if M~, ?< rcfcr to two that, modes of vibration of course to th same terminal (corrcsponding of -winch a rod is capable, then conditions) question on account

whcthcr

providcd The foiowcd, mcntd

?)t and attentive we ha.vc

Mt' bc rcader in

(Useront. will rctraecd was itsctf peroive thc that stops provcd in by iu the w)nch 1G2. proccss t))c It

mst fnndais the

fact

diicrcntial

quation

Tho thnt

ronder

ahonitl

obscrvn

t)nit

tho

eftscs

hcro

tho right-hand

monbt'r

of (;!) Vtuushcs,

provided

Rpoeificd t.)t)tt

MG pa.rticuJn.r

andV

<~ ~j t:Llll <~ Thoso towm'Js conditions incindo, for ~m<<j: < f/

~.E

its position witimut by n spring

of L'quili)')-iu)u inertia.

tho Ctlao of n rod whofiO end is urpod ittstfmco, to thc dispUtcc'ment, as Ly n terce pr~portional

164.]

CONJUGATE

PROPERTY.

209

originat ncctiun by~.

conthat ha.s the most nnmcdin.te MM'M!<MK<i!cquatio!) with tho conjug~tc If we dcnotc by M aud Sy propcrty.

and sired.

this

proof

is cvidentiy

as direct

and

gnral

as cou)d

bc du-

TIie whcn th

reader term

investigate tlie l'cprescnting

may

the

formula

rotatory thn.t

corresponding inertie is retnined.

to (6),

By !ne!ms of (G) we m!iy verify aro rca.1. For if 7~ were complex, function, fonction product
vanish,

thcn

i,8,

th

also, corresponding of the two functions, whcn Ititcgratcd in (3) w. and truc, and wc ?~' hc cannot

varies of n2 tlie admissible and 1t = a + !3 were a normal of u, would bc a normal conjugatc to tlie conjugate of ?~, and thon tlie being a. sum of squares, would not

If tica.lly

the at

samc, once

thc

quation the

hecotncs value of n~

Idcn-

Infcr

This mcthofl is, I bdicvo, t!no to roisson. R. 14

210
We th must take equation

LATERAL VIBRATIONS 0F BARS.

?~' equal to M + 8w, and trace th limiting as 87~ tends to vanish. In this way we find

[164.
form of

betwecn Now

the limits, whether an end be clamped, supported, ~V=0, ofa; at one end of the rod, or free,

M~"=0, and thus, if we take the origin

==~(~-2~V+~),(8). Th tion form of our integral is independent of th terminal If th end = b& free, M" and u"' vanish, condiand ac-

at x =0.

cordingly

is to say, for a rod with one end one-fourth of tbe terminal value, and l)e clamped, or free. supported,

that

free the that

me~n

value the

of u'

is

whether

other

end

1G4.]
Ag-a.in,
vanisL, a)id

VALUES
if
(8) wc suppose givc's

0F
that

INTEGRA/FED

SQUARES.
~t ==

211
n.nd M

th rod ia c){nnpcd

Since the

tins other

must end,

good wc sce tliat

huld

whatevcr for roJ,

be the one end

termina.l of which

condition is fixed

n.t and

theot.itcrfree,

thnt in t)iis case shewing thc c!ampe() end. TIie in thc a.!i!)cxed table diffrent cases.

M' at the

frec

end

is the

samc

as M"' a.t

gives

t)ic vahies

of four

times

th

mea.n

of M*

c!tunped,frpf. free,ft'eo clf~mped, c]!i)npcd supportcd, supported, snpported, supported ft-eo chmpL-d

M(ft'ccend),ot'M"'(cIu.mpedend) M'(ft-Rccnd) M' (clampcd 2~ end)

(supportcd end) = 2~" M" (freo end), or -2M'M'" (supported end) M"' (damped 2M'M'" (supported end), or end)

a. simpler free or a frec-frec

By assumes

th

introduction form. rod,

of In

these th

values case,

for T expression of a clampedfor example,

th

where 165. values

the

end

<c=~is

supposed method

to befrce. may be applied to investigate th

A similar of jM"~c, section and

In the derivation

of equation

(7) of the

WM assumed th truth of th preceding nothing beyond and since this quation is equally equation M""=M, true of any of th derived we are at liberty to replace M by M' or u". functions, Thus

142

212
ta.In c~es. For between

LATERAL

VIBRATIONS

OF

BARS.

[1G5.
in ail threo

tlie

limits,

since

tlie

tcrm

M' vanishcs

Il frec-frec

rod

of ~M' must bc cqn~I for, as wc shall sec, thc values or ngative at thc two ends. WhcUtcr u bc positive is positive. For a rod which is clamped at a: = 0 n.nd free at

and

opposite at a' =~, ')(t/

= l

We ])a.vc ah'cadysccn so that M"'=u/,

that

~"=~

a.nd it. will appca.r

( 173)

t)):).t

Il

rcsult

thf),t

wc sliall

have

occusion

to use latcr.

By n.pplying find

thc

same

quation

to tlie cva.inn.tion

of

~M'

wc

sinco

M'u" a.cd Mu'" tins

vanish. with (8) 1G4-, wc sec that

Comparing

whatcvcr

tlie

termina! result equation

conditions maybc arnvud

may

hc. dirccMyby intcgmting

Tho samc by pa.rts th

at more

1CG.]
16G.
normal

NORMAL

EQUATIONS.

213

We may
co-ordina.tcs.

now

form

th expression

for

V in tcrms

of tlie

If sion

thc

fonctions reducesto

:t

bc

thosc

propcr

to

:).

rod

frce

nt

~=

t)us

expres-

In any

case th

quations

of motion

arc

of tlie

form

is by dfinition the work and, since ~~t furcus during tlie dispin.ccmcnt 8~

done

by

the

Imprcsscd

if YpM~<; Le thc lateral If thcre be no impresscd

force
forces,

actmg the

on thceictncnt cquatiou reduces

of mass to

pax~c.

~+

n.s wc know 1C7. te

it ought Th

to do. of thc the terminal rduction values of of thc the Intgrais aud

signidc~ucc on be

~(~ its dcnvativcs !me rod of

dcpoidcnco may

function

in the normal at A-=~ vibrating A~ bu madc to the If a sm:di addition mode cxpressed by u. of a~ a function rod at ttic frec end, th form uf K (cons~ered with thc gmicral principlG but, l!i accordaucc ~) is ehanged, tlic period iV. ( SH). wc ma.y calcntatc iti Chaptcr CHtabii~hcd and free

fu'gmncnt. at x=(), chunpcd

p!a.ccd in To fix tlie

n clearcr light considur ideas,

by thc foHowing tho case of a

214
under th

LATERAL

VIBRATIONS

0F

BARS.

[167.
for the change

altcred

circumstances

wititout

aUowance

of type, if we are content to ncglect tbe square of thc change. In consquence of th strajghtness of thc rod :).t thc place where is made, the addition therc is no altration in th potentiel and therefore the altoration uf period energy, dpends eiitirely of ?'. on th variation This quantity ia incrcased in the ratio

which

is

also

tlie

ratio

in

wliicli

tlie

square

of thc

augmcutcd. Now, as wc sliall suc varies as f, and thet-cforc the change is in the ratio

tlie preseutly, in the square

is period actuat pcr!ot! of tlie period

A con)p:u'isou

uf th

two ratius M;'

shcws = -i.

tlint

ft(W~ inalsted

TIie but tgral

above

it serves

rcasoning at least to

a sunitar but it would tnay often require manncr, care to predict with certainty what atnonnt of discontnunty in the varicd be admitted without out of the range type might passiug uf the principle on which the argument The reader dpends. examine tlie case may, if lie p)cases, of whieh a small piece is Ititerpolated. 168. course nornial types, <neans bas In treating problems bceu to dtermine the to of a string Iti the jniddic

is susceptible. be treated in

upon as a dmonstration, tlie rduction of which ttie inexptain Other cases in winch sucli Intgra]~ occur

is not

the intgral formuim investigate by of which th particular solutions be conibined to may huit arbitrary initial 1 have prefen'ed circumstances. to follow a dinercnt to bring ordcr, t)ie bcttcr out the generality of th jnethud, w/~cA (/oes not depend M~o~ (t knowledge of the 7:o?'?~a~ yM/tc~'c~s. In pursuance of th same plan, 1 shali now investigate

viz. functions, and afterwards

to vibrations relating in th first place the functions represeuting

t!)o forms th

usnal of the normal

168.]
the conncction of th

INITIAL

CONDITIONS.

315
circumtreated

stiUlces, and solvc uiidcr th head of Strings. Th gcnend value

with th initial constants arbitrary to those oue or two problems analogous

of~

ma.y be written

formuloe

which

dtermine

the that

arbitrary we do not

constants need to prove by (1).

B,. analytically If a~ the

It must thc

be observed

of th expansion possibility solutions arc iucludcd, particular most general vibration possible, to represent Let position us of any now admissible suppose that initial th

expressed

th (1) necessarily represents and may therefore be adapted state. rod is originally in motion by rest, a blow that differs at in its which is, at from

and is set cquilibrium, to a small of it. portion lilitially, imparts velocity = 0, thc moment whcn tlie rod becomes free, and of one point (x = c). zero only in th ncighbourhood From (5) that (4) it appeurs that the coeiHcients

vanish,

and from

216

LATERAL

VIBRATIONS

0F

BARS.

[1G8.

CiUling ha.vc

~~<u~,

thc

w!)ule

momcntum

of

tbc

blow,

Y, wc

If poncnts, prcsunt uf101.

thc

blow tha.t

bc app)ied

at

a no(te

of onc of thc

normal motion.
investigation

comTlie

is in th rcsutting conponcnt missing ca.)cu!atiun is but a. pai'Licular c~c of thu

we may take the case of a bar, examplG which is initially at rcst but dcHected from its natural position force acting hy a latral at .'c=c. Undor thse circumstances the coefficients B vanish, and tlie others arc given by (4), 1G8. Now

ICf).

~a

another

!)i which bctwccu

tit tlie

tenus inuits;

frce by

from t))e

arc iutcgml sigii nature of th c:).sc

thc

to 'bc satisdes

takcn tlie

169.]
same vanisli tcrminal a.t both conditions limits. bc to thc .cicmeut har tri vos

SPECIAL CASES.
<m docs If yi~c, tlie th external cq~a.tion and thus force of a.ll thse initially

217
tcrms applied ci' tlie

equilibrium

If

wc

now

suppose

thn.t thc

thc

initia

dispiaccmcnt

is ut' t))c

duc

to

a. force applicd a; = c. wc tiave

in

immdiate

mjighbuurhuod

punit

a.nd for tlie

complte

value

of y at time

t,

made we have not hitherto expression but at th ends, as to the couttitions assumptiuns any special to ttte case of fi ba.t' which is c!cnnpcd curscivus if we now confine at a; = 0 aud irec at x = l, ve may replace In Jenving the above

Ifwc

suppose

furthcr end,

that

the force c=

to whk-h wc get

thc

Initial

dcHcetio))

is duc

acts at th

so that

Whcn tncnt.

t=0, Iti cases

this

cquadon

nuist

represent

t)'c tnay terni

initiid pi-cscnb of which

of this

Idnd

di~culty

dispjacc:itMu[f as

satisfics for the series, cvery to !)0\v it is possible Iti au initial displacement condition th y"' = 0, to rcprcs~nt after Th iact is, that triple is violated. tins condition which no longer tlic series converges wiHt respect to diH'crentiation of y" is not to be ~rrived the value and accurdingly for a~, the terms first and summing the diHerentia.tioas at hy making

218
aftcrwards. <ve cousiJer

LATERAL

VIBRATIONS

0F

BARS.

[1G9.
be a.ppa,rent if

Th a point

truth
distant

of

tins

Rta.tement the end,

will and

dl from

replace

For the

th solution is referrecl Thc cases forms arc

of tlic

reader 170.

prsent to 101. normal

probtcm

by normal

co-ordiiiatcs

of tlie

functions

ni the thu

varions ratios

ticutar

to bc obtained

four constants

iu thc gnrt

by deterimuing solution of

p!u'uf thc

If for thc heputintothofonn

sako

of brcvity

be written

for

th solution

may

cosh x and th quations

sinh x arc tlie hypcrbollc

cosine and

sine of x, defined

by

1 hve a special

foltowed symbol

thc usual might vcry

notation, though weH be dispcnsed sinha;=-.t'siu

introduction with, since .(3)

thc

of

cosha'==cos/.c,

w]~t-e t== y- 1, and then tI)G conncctioti between tlie formuh~ of circular and hypcrbo)ic wou)d Le moi-c apparent. Th tngttnoniet.ry ruics for diiTurcntiation arc cxprcs.sed in ttic cquatiuus

In diicrentlating (1) any number oftimes, funetions as thcrc occur are pound contmuaUy one of them which does not vanish with on!y wbose value is thon 2.

the same

four com-

Th reproduced. is cos a;' + cosh ic,

170.1

NORMAL

FUNCTIONS

FOR

FREE-FREE

BAR.

219

Let

us take

~h'st the

case

in which

both

cnJs

are free.

Sincc

values of ?/t. whose roots are tttc admissible is the quation C given in (5) ~rc e(tual, of tlie two ratios If (7) be satisfied, multiia (4-). The constaut substituted of titcm ma.ybe :m(l either wc have for thc !iorm:d function plier bciug omittud, This

171. a velocity

The

frcqucncy

of th on

vibration the material

is

~M~,

in

whiuh th as bar

& is is

formcd, and mode radius

only dcpcnding a-nd Ht. is an abstract of

of which

number.

Hence varies

for a given clirectly

material Kthc to th

tlie vibration frcqucncy about of tlie section of gyration

an axis perpendicular

~0

LATERAL

VIBRATIONS

0F

DARS.

fl~I.

plane of bendingand as thc square of the Jcngtit. Thcsc iin'erseiy rusults hve bccu anticip~tcd might f'rum dimcnby thc argument tf if worc considrt Stons, that t))C frcqnency is Mcccssari)y dctci-tninc-d Ly tl.e v.d)!e widt of th~t of ~tho togeUtcr uuly qu.mtity on sj~cc, timc and mass, which occurs in (]cpendh)g t)to diH'crcntiid cqn~ion. a bar be given, If cvcrytiting eonccnnng its absolutc cxcept tiic frettuency m:~It.udc, vanes as invcrscly thc iincar dimension. Thcsc I~ws find an In thc case of tuning Important application furks, w))osc prongs vibratc fm rods, Hxcd at t))c ends wherc thcy juin the stal!~ and frce at tiie othcr cuds. Thusthc pcriod uf vibration of furks of t.hc samo tnatcrinl and shapc vancs as thu lincar dimension. Th period will Le approximatdy of th indcpcndent thickncss to th plane of bending, po-pG!)dicn)ar but will vary inthc thickness in thc plane vcr.scly with of WIien thc bcnding. tliiekncss is givcn, tlic penod is as thc square! of t]ie length. In ordcr to ]owcr th pitch of a fork we for tonporary jnay, load thc cnd.s of th purposcs, prongs witli soft wax, or file away thc mtal near thu base, thc sprinn'. thcrcby To raisu wcakcnitig thc pitch, thc cuds of tit prongs, which act by inertie may bc filod. Thc which t)ic
tunni~

value it

of b attains to Le
of bt-~s

its maximum 5237 mutt-cs in about th

woutd bc a.bout

in

tho

case

amouuts would
n~tc miLtcrial

about less stcd.

vcloclty fork ii'thc 172.

th~ti

per second. ratio 1'5 1, so that a a. Hfth lower iu pitch

of steel, for For b'mss

were for
dcrived

TIie
un!ncdi)ttu]y

solution

the

ense at

may

bc

from

w!)cn buth thc prcccding both ends

~rcutiation. ditions

Since

y satisnc~

arc dampcd difby a double tiic terminal con-

ends

~'hich

arc thc

conditions

dtRerctit.iid
oniitting

eqtmtiot)
cunst:u]t

for :L chunpcd is a]so s~tisticd

end.

I~torcovcr
Titu.'j wo

t))C gnerai
may

multiplia-,

hy y". as bcfur~,

tal,

173.] while

NORMAL 7K is given

FUNCTIONS by thc same

cos H:

FOR quation
cosh

CLAMPED-FREE as hcforc,

BAR.

221

namely,

??t=l.(2).

We

conctudc

that

the component

tones

hvethc

samc

pitch

in tho

two cases. In the each case therc of arc amt systoms its dcrivativcs. four of points determincd by thore

evanesccncu

is a nodc ~hcrc wlicro mcnt y"

Vimishes,:). place uf

~anisites,
vnnis!)RH, maximum

vanislies, WIien a. loop, or place of maximum displaccn. point of inftection and whcru
curvaturc. Whcre thercaru in th

ni-st

case arc and

(frec-fr) points of iuncction in thc second (chunpcd-chunpcd) vice ~er~, points of inftcctiou rod correspond

and

of maximum and
maximum

culture, loops
curvature

there for

nodcs and of to nodcs

rcspcctivcly; rod whoso

a douhly-clampud ends arc free. 173. a;=(), 170

a;=0,

aud loops

ofa

We will and frec at

now consider a;=~.

th

vibrations to the in virtuo

of a rod clamped guncral of the intgral conditions

at (1) at

we sec that
so =~ that (cos;r'

Rcvcrting ~1 and C' vanish

D

cosh~)

(sin

.r

sinh

~')

.(1).

The

remaining

conditions

at x = givc

.D ( cos ?)!. + cosh ?M) + D (sin ni + sinh w) = 0 J? sin ?~ + sinh ?M) + D (cos ?K + eosli ~t) = 0 )} whence,omitting multiplier, ) \f u (sin + si n = (sm ne sinh?~) M+ ~cos t ( (cos?H+ or M = (cos Mt + cosh -;cos ( Ma: 6 t i ~~) cosh tJ sinh ??!.v] J .(3), ( cosh?~wsm srnii the constant

cosh t

i ") J "~1 (2),

( smm-Sllll1n + smhw~sm +(smmwherc M: must bc a root of lf

cos M cos!) 'w -)-1 = Thc thus httcd pcriods dincrent to, thosu of the componcnt
as

0 th see prsent presently, proLIcm

(4). arc rc-

toncs
wc shall

in

from, though, of a rod both

wtiose

ends

are clamped,

nearly or frcc.

222

LATERAL

VIBRATIONS

OF

BARS.

[173.

If th sult At The free (!) .u=0,

value satisfies ,t", <

of !( in (2) or (3) be diftcrentiatcd twice, the rcuf course t!ic fundamcntal diffcrential cqua.tion. ~.c" )i"ani.sh, but nt.<;=~ M"a.nd -r~.r va.nish.

function

?<" is therefore

at 0, proving that th in th origina.1 curve :u'c at th samc curv~ture end, as thc nodes and loops respcctiydy clampcd end. 174'. rithin, Ta.lun~ th nrst In dcfault admissible ttic cquation of tn.htcs vaincs of tho of
M

to a. rod clumpcd ft.t and applicable of inncction and of maximum points distances arc from from tiic the free

hyperLoHc
may bc

cosine

ca)cu!atcd

nr its !ogaas follows.

we ~(2t

see +1)

tha.t

~t,

when
au

Jfu'gc,
intcgcr.

Tr, i being

a-pproximate If we assume

must

in

value

to

;8 will bo positive Substituting

and in (1),

comp~ra.tivcly wc find

small

in magnitude.

eot~=~=~

an

quation

which

may bc solved and result e is in

by successive ascending

approximation powers of th

aftcr small

cxpamHng <;uantity

tan~ The /3.

which

is sufficiently

accur~te,

cven

whcn

t= I.

By cn.lcula.tion. /3, = -OI79CGG ~t. ~g, ttie sries -0003228 still + 0000082 -0000002 = -017C518.

arc found gives

Thia

~3 correctly
ia aomewhat

more easily. th first term of Aftcr as far as six significant Th ~g~ires.

Himi!ar to that adopted by Strehikp.

prncoRs

174.]
table /3, and contains tlie value the

CALCULATION
value of ~3,
th

OF PERIODS.
angle whoso circu]n.r

223
measurc on. is

of siu ~/3,

which will be required -F'ee-Free j~~)'.

fm'ther

.q
1 2 3 4 5 10'' x -17G518

oxproHsed in dcgroea, minutcH.tmdtiCcondt). 1" 0' 40"-9.t 2'40"-2G99 C"-92029 -2MOG2 -0129237 10-'

~3 ~2' x -88258

10-777010 10-" x-335505 10-'x-144989 10-'x-C2C55G

10-'x-38850 10-1G775 10-" x-72494 10"'x-31328

Th

values

of

which

satisfy

(1) are + /3, = /3, = + 4-7300408 7-8532046

M, = M, =

4-7123890 7-8539816

= 10-9055743 = 14-137~1669 = 17-2787596

= 10-995C07S = 14-1371655

+ /3, = 17'2787596

e"=cot~=~e~ where however this a' = e~

From as that In fact

it appea-rs that th series of /3, though the corresponding ~=~, a,=/3,cf~=/3. furthcr to

of values sufHxes

of a is the are not the

samo same.

so that however

we the

ha-ve series

nothing

calculate

than

(4) is not

sufHcientIy

convergent.

c(~ for which Th value

Thia connexion noticed.

between a and ~3does not a.ppeM to have been hitheto

224

LATERAL

VIBRATIONS

0F

BARS.

[174.

of a, may

bu obtaincd log~cot

by trial a,

and crror

from

the quation

'C821S82

--t342:)-<-48 a, = 0,

a.nd will

bc found

tu bc a.=-304.3077.

Another givcn

method

by wbieh

/)!, may

bc obtamcd

(Hrcct)y

will

bc

prescntty. vaincs of ?~ wtiich M, w, M, = = = = 7)!, satisfy (5), arc + a, = = + aa = 1-S7.~04 4-C94737 7'85-)<758

Thc

1-57079M 4'7123890 7-85M.S1C 10'!)!)55743

= ]0'9!).554.1 + a6 = 14'137JU8 = 17'278759, 1

= 14'137tCC9

7M, = 17-278759G aftcr which

7H = ~(2t are proporThc frcqncncics l)'7r Rensibly. tones tional to ?M", and ~re thcrtifurc for the highcr ncariy in thc of thc squares ra.tio ofthe odd nmnbers. in thc ca~c of Howcvcr, of vcry high order, ovcrtcnes thc pitch may bo stightiydisturbed whosc effect is Itcre ncglectcd. by thc rotatory inertia, 175. Since th arc satisfy
cosM cosh m = l.(l),

componcnt nccessa.rily

vibrations of tlic

to dissipation, of Mt", which

of a system, Dot subject harmonie type, n!! the values

must w, hve

be w

reah 1,

We sce furthor ??i. 1.

that, Hence,

if M bc taking nrst

a root, th

so arc lower sigt],

also wo

If we takc cfHcicuts,

thc we gct

logarithms

of both

sidus,

cxpand,

and

cquate

co-

This

is for a clatnpcd-frcc

rnd.

175.]

COMPARISO~

0F

l'J/J'CH.

225

F''om thc known value witli tlie aid ofapproximate 2~

~nd M~

of S?)~, values

th value of?y~, w,

of )?~may Le dcrivbd Wc fmd

=-0065t7C2J,
= -OOOOO.i-237

~=-000000069 ~=-000000005. whencc j~iving In likc manner, 1 '-~+-=~)('

wliciice whcnce 9It

w,='OOG5i.33)0 = '187510;'), if but.)) ends

4 + l 7)r

n.s befo)-(\ or free,

of t))u Lar bc el.nnpcd ~W. I ~n n .Cc.

(4),

= ~~a' zero The :2,tJ ) J value

whct'e of nf entil-se crmrsc &c, %vliet-e of??{.

t))p ,;Ill)lnl~itioli tilt, sumDin.tion

Is is

exclu-

sive

ofthc

17C. w". may octave.

of thc frcq~cncics Th interval between any bc expressed con'vcnientiy Tins is effected logarithm r4G2<) 2't3.')8 ~'1590 3-7382, ~c. relates and but that to the free tlie by dividing of 2. Thc

are proportional to tonc and the gra.vcst of th sries and fractions h) octaves of an the diffurencc rcsults 2'C478 8 4-1:~2 .IO!}G6 ~-8288, the toncs &e. of a rod other. hoth whose of a the <if thc logarithms are as fotlows

sorics

of toncs

of w' hy th

wliere ends rod

the first

column

are clamped, at clampcd column

or free; ottc end we find

second at th first

column

to the case Thus is 2'()-t78 from

second

overtone

octaves

than th gravest tone. The fi-actioiial part may be rcduced higher to mean semitones The interval i.s then by multiplication by 12. It will be seen that two octaves + 7'7736 mean scmitons. th rise of pitch is inuch more rapid than iu th case of strings. be clamped at one end and free at th other, th pitch of the gravest tone is 2 (log 4'7300 log 1'87.51) log 2 or 2-G698 octaves lower than if both ends were clamped, or both free. R. 15 If a rod

22G

LATERAL

VJBRATIONS

OF

BARS.

[177.

177. rod

In ordcr

to examine transfbrm nutncrieal free.

more the

closely

the

curve for

in which M into thc

thc form case n,

wc will vibmtes, more convcnicnt for when both ends arc

expression

fu'st odeuladou, takin~ Sinco w=~(2t+l)7r(l)'/3,

and cosM=sin/3, siu?~=cos'7rxeos~3; thoreforc, root of cos M ccsh ?/=!, ccsh Mt = coscc /?. AlijO Hinh" w = coHii" ~t
or, smcc cot/3 is positive, sit)h~ =

bcing

1 = tau2 ut = cot' ~3,

cot/3.

Thus
slM?):8[nh)~ lcos!'7rsin/3

cosMtcosh?~

eus/3
(cos ~/3 Hit) cos ~/3) 't'Tr sin (COS ~/3 ~/3) + COS ZTr Si)! ~)

(COS

COStTT

ces ces

~/3

cos cos

tTr 'Tr

sin + sin

A/3

Wc

may

thcrcfore

take,

omittiog

the

constant

nudtiplio',

SiIl =~cos<7r.s.n~ Mj

(~~

7T

i~f/3)

-~+(-])'~ t~ .(~. the factor ~/2, an(tput~=l,wc

+SHl~e'-COSZ7TCOS~C'' If may whcrc = cos '-n-sin {;);

!og7~= ~cloge+Iogsin~-logys Mi~ e + log cos

wc ta.ko

furthcr

throw

out

M =~+~+7~7r + ~( 1//3J
'(2),

!og

~,=

log

log

~/3

} of i an'!

from

which

7~ may

be ealculated

for dUTercut

values

177.] At thc

GHAVEST ecutro of

MODE thc bar,

FOR. = Whcn

PREE-PREE

BAR.

227

are numcrically and i is ~fc~ thcso tcrms ea.ncel. cqual i)i ~irtue of e'" = cot ~3. is cqn:~ which to xcrowhcn For.F~weha.vc ~=(-l)'siu~7r, i is evcn, tuni to i 1 whoi i is odd. WItcn is even, t)io'cfoi'c, <hc! sumof thc threctcrms'v:mishcs, and thorc is accordin~y n, nodc in tlic mi<!d[c. Whcn
(since Is

= 0, M reduccs
a.Iw!).ys smal!)

tn shows

2 (- l)'sin (.~ 7r that for no vfttuc (hcl(),

(- 1)'/3}, winch of i is t)icre a. ut, th of in

nndo at thc

end.

bc gcnt)y whilo centre) vtn'ying points its length !u-c damped wit)i th nngcr.s, n.n unu.su:d dcaducss thc souud will bc uoticcd, as the end is cluscly approacttcd. 178. F~n~'o): Wc will w~/t. now t:).kc somc p:u'ticn!ar cuses.

If a long ]):u' of steel with n, ita.mtncr t~pped

fur exempte,

<wo HOf~M. is th

i = 1.
~ravcst

If -t'= 1, th vibration Our fonnuhe bueotnc = h'g log from whicit 7~ = = sin

of which

the

rod is capa-btc.

(270 + 1" C' 40" '94)

-M"

30' 2()"'4.7}}

2 054231 a; + 3-7!)52301 2-054231 is calculatcd to 'OU, '05, a; + 1-8494G81, fuDowlng '10, &c. titc table, giving thc values of

M for a; equal Thc values

of M :M('~) for thc intcrmcdiatc values ofa; (in tlic last column) werc found by iutcrpoht.tion formulK. If o, ~,?', N, t be six consccntivc that intcrmcdiatc terms, between aud r is

228

LATERAL

VIBRATIONS

0F

BABS.

[i78.

I~

I n

7~ I~a

?;. at.

~c M:~(-5) 7c(')

-000 '025 -050 '075 -100 '075. '125 '150 '17~ '200 '225 '250 '275 '300 '325 '350 '375 '400 '425 -450 '-175 '500

+-7133200 -5292548 -3157243 +-084GIGC --1512020 -3786027

+-OOG3408 -0079059 'O1001o3 -0100153 -012G874 -01G072C -0203G09

+-7070793 -5581572 -440GOC5 '~140G005 -3478031 -2745503 -2IG7256 -1710798 -1350477 7 -10GG045 -0841519 -OGG4282

!+l.42GC401'+l-G45219 1-45417G 1.0953179 I '7GGJ401 -7GG3401 -4451071 + -1394209 -14151C2 -3880523 -5909G08 -7422059 -8355740 -8G7I433 + I-2G3134 r0721<!3 1:0721G2 -8837528 -G9<!9004 -5133028 -3341(!25 -1G07819 -0054711 -1G31982 -3109982 -44750GG -5714137 -GR15Q32 -8559210 -9184491 -9G35940 -9908730 -1-0000000

-5849255

-0257934 -0326753

-7586838

.7766G2!) -8902038 -0413934 -9721G35 -0524376 G +-OGG4285 -1-000000

Since middie ~='5.

th Thc

vibration rod, curve

of th

is symmctnca,! with respect it is unneccssary to continue the table itself is shewa in Hg. 28.

curvc

to t)ie bcyond

To Hnd th position

of tlie

node,

we bave

by

interpolation

~1 G(i2530

178.] which distant

FREE-FREE is th fraction from th

BAR of tlie end.

WITH whole

THREE Icngth by

NODES. which tlie uodc

229 is

Huarer

Vt&)Yt<MM ?~~A </<re nodes. FI log~= log (- F,) =

=s:n

i = 2. .B-4.5" +1' 20"-135) }

( (450

2'40"-27)

3'410604.c+4'438881G 3-410G04 +1-8494850.

iC

~M(0)

M:-t(0)

-000 -025 -050 -075 -100 -125 -150 -175 -200 -225

-l'OOOO -8040 -G079 -41477 -2274 + -0~87 -1175 -2G72 -3973 -5037

-250 0 -275 -3O -325 -350 -375 -j00 -425 5 -450 -475 -500

+-5847 -6374= -6620 -6569 -6245 -5653 -4830 -3805 -26277 -1340 -0000

th values this table, as in th prcecding, of !( were calcu]:t.tcd directiy for x = -000, '050, '100 &e., and intcrpotated for thc of thc nodc tlie table gives ititcrmediate values. For th position a; ='132. from th above by ordinary ititerpolatioM C:T.lculatiug wc fiud formul, In ~(-1321) =--000076,

M(-1322)=+-OOU88, x = '132108, with the result obta-ined agreeiug The place of maximum excursion from may be found function. We get \vhen.ce ('3083) whence Hcnce the value excursion == + -00~6077, u' (-308373) (.~081) = 0. then attains th = by Strehike. the derived

-0002227,

it is a maximum, when a; = -308373 be observed, is mnch -6636, which, it should at the end.

less than

230

LATERAL

VIBRATIONS

0F

BAHS.

[178.

Thc

curvc

is s!)wn

in fib. 2~). Fig.Si).

r~t'tt~M

M~/t ybtO' 7'~ =

)!O~CN.

i = 3. 45" 3"-4G],

sm [ (G30" + G")2) 4-33~ 4-77.'i!~2 .e + 5-0741.~7, + I-S-~4850.

if~-

Io~ 7~ =
From

th

uodcs

~(())=1'4M24., are ruadi)y foumt u (-3558) =

t))is

M (.~)= 1-00570. by trial and crror.

T!ic TIms

positions

of

-C()()037 Thc \t)ut;

M (-3-')59) = + -001047, of :r ibr thc nodc nca.r the (nid

whcncc is -09~ Thc

M (-35.~S03) (Scebeck).

= 0.

of the loop i.s he.st fuund from t!te dc'rived position function. It thut w)ju!i a;=':22UO, ard thc!i ~'=0, ~ppc~rs Tibre is a)so a tuop at thu centre, M =34-9. whcre I)o\vVt.;r
t!tc excursion is not

so grc:t.t

as at thu

two

uUters.

of thc bar 7'~ :H)d ~u'c n)t)]icric:d]y lu thc nci~hbom-hood of t])c )))i<!([)c-, 7'~ is L'vidcntly eqn:U. vury bc sni:)]!, if fur tbu njdcs modurn.tciy ~rc~t, fmd thus t))C c~uation rcducus to approximateiy

Wu sa.w t))!tt at thc

centre

?!. bcing f),u int~cr. thc ru(.),~nd rcptacc Hnd

If

wu

tr~tisf~nn

thc

ot'igiu vainc

to thc K2~+])

centre

of

7?t by

its

approximate

Tr wc

178.]

GRAVEST

MODE

FOR

CLAMPED-FREE

DAR.

231

shcwing

tha.t

ncM'

thc

middic bccu

of thc consceutivo

bar

thc nodcs

spac~d, thc intm'vit.t bctwecn Tins t))corct,icn.t rusult lias Strchtkc and Lissajous. of

nodes by
tit

are uniformiy bcing 2~ (2t+ 1).

mca.surcmcuts of

verifiud

F(.'t' mutliods thc

n.pproxiin:(.tio)i

npp)ic:)bic:

to

thc :uid

nudcs to to

nc~r t)tc

th:n) 3, thc cud.s, whcn i is gre~tur mcinnir by Scubcck :di'eu.dy tnoutioued ~tcu:cs (p. 194'). 179. clamped ~'=~+7~+7~
Thc ca.lculn.tions

l'cadcr 160,

is rcf'~rd

]3on~in'!t

at

onu

end

for tho case of a. bar n.rc vcry simitar n.t thc uthcr. If to: a.u.d aud frcc in gcncrui

wc hve

Ift= cm'vu

],

we

obtain

for thc

culculation

of tlie

gravest

vibratiou-

Thse

givc

ou

ealcut.i.ti'ti

( 0) = -OOUOOO, ~(-2) =-10297-

~( ~(

-G)= -7-t3~2, -8)=riCUO:32,

(.-t) = -370G25, frutn which fig. 31 was cot~tructcd.

F(l-())=l-G1222-t,

232 Th distances

LA.TEHAL

VtBRATIONS i'rom are

0F

J3ARS. cud in tlie case

[170. of a

of tlie nodus eud

tlie free

rod clamped 2"tonc

~'i.(Htu

at the other

given

by Secbcck

aud by Donkin.

-22G1.
-132<, -4ij!')i).

4.to!tu ~hm~ 4~ "Th that 4/-3

-0!)-t-4.,

-3-').')8,

-04.3!). ~3 4~-7~)75 t,

last

row

in this

table as the tin-ce

must distM)cc aud

be

understood of thej)' last

as meanincr uodc from th

may

bc takcn

froc (.'nd, cxecpt Wlmn ncarcr end 1" tone 2'tonc-1321 3"' tone i"' ~tone't-<+2 Th buth are

for t!tu first ends

thc

two nodes." uodes from the

are ft-ce, tlie

distances

of the

'2242. -a. -0!)44 '3.')58. ~i~ ~t'+2 for -3 4t'+2' rod (corresponding by SccLeck to

4t+2

of inUcction points thc nodcs of a chunpcd-dampcd

a h-ec-frcc rod) arc atso

givcn

1~

point.

2"~

point.

t''point.

l~tf)no
2"tone. -f)f)O

No

inaecdon

point,

3'tone tl tone i ~"tone..

-03 i~ 4t+2 S.9!)!)3 -in-2 LI 4.+1 ~+~ 1

Exccpt responding
uceur m close

in

th

case

of th th

extrme nodes

nodes

(\vh!ch

have

uo coralw~ys

infieettou-point),
proximity.

:md InHection-poiuts

180. ~o~eJ

Ttiu case whcre onc eud of:). rod is ft-ce and dous u~t ubcd an indcpcndent investigation,

the other as it

s~may be

180.]
rufcrrcd to that that is, with anode y aud v" vanish, winch end. In hkc nianner are the saine as tliose

POSITION 0F
of a rod with in are both ends

NODES.
free M'M~ For attitc thc conditions of uf a rod

233
in an e~?t wof~, node central

themiddie.

precisely the vibrations a central six arc

for a supportcd rod a clamped-supportcd both wliosc ends are

of one-haf

c)amped,

vibrating The whcn last both

with of tlic ends

nodc. oi' tenninal conditions we

181. occtu's sec that

combinations

tlie conditions

supported. = at x 0, give

to (1) 170, Refcrring so that ~1 = 0, -D = 0 D) sinh C' D = 0, and

= (<7 + D) sin .e' + (C Since

Hencc

M and
the

M" vanish
solution is

when

a:' =

sin Ht = 0.

'J'TT.'r y=sin wltcrc course It intcgcr. and be prcnxcd, tlie i is an -cos~

An

~TT~X~

<

(1),
constant

arbitrary

a constant normal two

may curves fixed

be addcd arc tlie points,

inultiphcr to t. sanie
but the

may

of

that appcars of a string stretchcd toncs cf i. is altogcther Th uodes arc also

between

as in thc case
scqncncc of

bctwuen

(which
tances

as tlie square tlie frcqucncy dirt'crcnt, varying and thu loops and concide, InnccLton-points biscct thc disof maximum the points curvature)
uodes.

thc

182. tlie

Th

theory

of a vibrating
that th natura]

rod mn.y be appHcd

periods of a system

gcnera.1 principle maximuni-ininilnum

to illtistrate fulfil the

of th natural and that the greatest condition, of can that be obtained exceeds by a variation any periods rod is of a clamped-free curve tliat th vibration type. Suppose itself if dcnected by a force that in whieh thc rod would dispose of th curve may be The quation at its free extrcrnity. appHcd taken to bc y=-3~+~, which 0, and satisfics at = 0 throughout, b and makes of th y and J rod y vanish J at time (1), C~cos~X, while th t be at

<<

Ttius, ~=

if thc configuration (-3~+~) is by (1) cos~ ].61,

thc

potcntial

cncrgy

234

LATERAL

VIBRATIONS

OFDARS.

[182. 2] 40 7 9

J" kinetic

cncrgy (U)c

I.s truc

33

n l7

~sin' v~tuc

and 1] tlic gravest

thas

~=~ tonc) is cqual to

~ow

of ;) fur ~(~~J.

suthat

shewing

that

thc

coniparativcfy cal type. Jt

question violtes

gravest IitUc)!owerthan t)iatca!culated is to bc observed tbat thc

thc terminal

i-cal

pitch

of

tho

tonc from

is the

rather

(but in

I)ypotheti-

however, for the unti:d Jnent Wc I~tioa

(tocs
assumed

not

intcrfcro

typu

couf~m-atiou of pcriods. )nny cxpcct on thc cnrvc

of' ti)e pnncipi~ application niny bu :my whicii wouid bu admissibie as an but it tends to provcnt a very dose ngrcc-

condition with (hu

y"

hypothctic:d type = 0. This circumstancc,

a bottur
in whici)

by a force at somo litttc (ti.stancu actiug frutn thu frcc whicti c-nd, butwcen and the point of action of the force bo (.c = c) thc rod would and tbcrcforc witiiout strai~ht, Thns putential cncr~y. potential Ti)C kinetie t))c ~'atuu ofy.
From 0 to c y = :}~ + i

approxitnatiot), thu rod wou)d

ifwc

found

our calcu-

bc d~flectcd

eno-gy can

= (J y~M~

cos' by
intgration

cno-~y

bc rcadify

found

from

amt as

from
may

c to L
bc sccn

y = frutn at :c= the c.

(c

3.<'), that yand y' nn)St not

sudd<jn)y

change

considration Th rcmt)t. is

kinctic \yhcncc

cnc.rgy

sin'

(~-

r) (.' + 3f-)1

~=~[~+~]The jnaxinmm

"12

70 '3.1 Ga vainc hf

`~

(c2

e3Gt)

of thc force application second nornud vibration. compuncnt a result wllich is tw fngh in the o

occnr 1-~wiM is ill thu ueighbour)tood

wttcn of the

t)tc

of point nodc of tl)c

If' wc takc m~hicat scatc

c =~, ~vo obtain i by thc intun-a)

182.] cxprcsscd t,he trut)h of

a. vibr~tin~

LOADEDE~D. by th ra.tio 1 This cxampio '9977,

235

th solution Thc type

nea.r a.nd is ~ccordingly cxtremdy may givc un idca. how uciu'ty thc pci'iod rncans without by simple systort may bu catcntittcd or tra-uscuudenttd uf diHurcMti:d cqu:),tious. of
vibration

~ssumcd by a. bar whicii that tbe lu.td J/n.t its frec end, providcd Wc sliould bu ]mg!ccted. h:n'c, in i'act, F= sothat Evcn if thc i))Grti!i of tbe Cf/~N~ eus'
:<

just cousidered is itscif dcvoid

wout<t

be tliat

but of inertie inurtin ofJ/could rotatury

actua.ily can'ics :t

7' = 2~ .(.<). bc tid~ not titc

si)i'

V~ bar

eomp:u'isoM with jV,v'c may c'idcutation :m appruxijaatu

still

attogcthcr saniu typu

in n~gligibic as tticbasi.sut'

th:Lt is, J/ is to bc incrcascd Mincu titis rcsuk thu )'od. hot ([m'ur nmch (ruiu t!ic

of tLc mass of by n.hont onc quartcr is mfimt.e, atld dous is accm'!).k! whcn rccvun whcu~V=0, trnt)), Itrn:).ybu
a.pj'roxhn~tiun.

!).s a.u a.pplic:).b!u ~n.rdud as gutiuraHy will ahv~ys Le on tlie sidu of cstimatm~

Th tuo liigL. of ~f could

cn'or

ttiu pitch Inortia

183. bcjustiiicd

But

thc

ncglect

of th

It is as thc ordi)i:u'y of cxpenmoYt. ~ot to construct,a.c:).sc m whie!) tlic inertia unsyto Im:)ginu, thou~h s))un)(i bL; with thc iucrtia of in of translatioji ncgligIDc comparisou midd' rotation, If both that distinct as t)~ kinds aud opposite of incrtia. uxtrutne in thu wtuch !na.ss ~f bas just bccn considcrcd. cven thougli bp iuctudcd, two th systum possesscs

rotatory couditi~us

not

of t)ic l):ti' bc

nc~jectcd

indupendoit dnote thc

ft!to~ctiicr, of vibration. po'iods vaincs of and

Lct z and cquatiou

ut .B=

Then

tLe

of thc cm'vc

of thu b:).r is

~+

S3G
and

LATERAL

VIBRATIONS

0F

UARS.

[l83.

whileforthkIneUcctiu'gy ~=~+ If~ be the nutius of gyration L~ ofJLTabout .(2), an axis pcrpendicular J to

t!)ep)an(;!ui'vibnLtioti. Tbc equa.tions of motiou are theruforc

whciicc,

if z and

vary

as cos j;)<, wc find

con'cspo)iding If back wc

to tlic

two penods,

which iucrtia

arc aiways by putting

difci'cnt. /e'=0, we fall

thc rotatory negluct rcsuit on our prcvious oT is thon sma.

3f7~"M ~f

Tite

ot!)cr If ?'

value bc

of~ merc!y

infinite. so tliat its Iti~Itcr powcrs may be ueg-

lectcd,

If on the vented, vcntud,

other

hand

A:" be

vcry

grt, <~M

so tha.t rotation

is pre-

12<7A'~ ~=-77rth lattcr of which

or

tlie is prcvc!'t.c<), at a!). inertia rotatory

is vcry sma]]. pitch is an octave

T))cse

It

appc~rs thn.t when iu~Itci' than if therc

might

rotation were no derivcd

cundusions

also

be

183.]
dit'cctty tlieii from tlie

EFFECT 0F ADDITIONS.
diH'crentiat quations; for if/c'=~,

237
0=0,a.nc

butif/<=0, thatcase

~=~

by

thc

second

of

quations

(3),

and

in

If any addition of vibration is prohjnged. nrst would quires In clumpcd motion. but as thc of th The that thc ]t0 altration

184.

to a bar bc made at thc end, th periocl If tlie encl in question bc frce, suppose inertia. Since thcrc would bo pice addcd is wit)iout in eithcr tho potcntia! th pitch or kinetic nergies, as the a.dditiona.t part a.c-

be nncliangcd but in proportion the pitcli Mis ( 8S). inertia,

the sa.mo

of a. har bcyond a way :), smiUL conthiun.tiun end wonid hc wiLhout no nu'ect, ns it wou)d ac()ui)'c No change will cusue if tlie ncw end bc a.tso c):).mpcd chunping in thc is rc!a.xcd, potential od thc pitch faits, In consquence cucrgy of a givcn dutormation. is not

first diminution case of a

Lct tlie quitc so simptc. and let tlie added is at original piccu whieh to hve no incrtia., bc ~t/?. thc end ~1 is nrstsupposed InitiaNy stin'fixed, or held, if we )ikc so to l'cgiu'd it, by a spring of inrinitc this spring, which )ias no ino'tia,, is graduaHy ncss. Suppose tbat this of thc rclaxod. ncw end During proccss thc motion supportai tjn(L of th rod bc diminishcs, this During and at a certain proccss tlie pitth and to become nxcd, snpposed anothcr f:d! of pitcli, to Le further cntails inertia. 18.5. treated thcre Thc by the adoptcd case gencr:d of a rocl method
whieh is not quitc utufonn

point falls.

of relaxation, -D cornes to rcst. now at rest, may bc 7~, being the abolition of th spring at ~1 increased as ~J3 acqnircs

may

bc

of

90,

We

))ave

in

thc

notation

238
whcucc,

LATERAL VIBRATIONS OF J!AHS.

P,. bc'iog t))C uncu)-)-cct,cd value oi'

[185.

For

examplu,

if the

rod bc e!:unpu(!

at 0 and

fr

:tt

Thc

samc

fonnu!~

appiics

to a doubly (~V is thus

frcu bar. givctt Ly

T)ie e~uct

ofa. smaH lo~d

who-c t!iocn<],it,s ratio ~+~J/: ~8G. corroctio)!

dnotes cficct

tl.c

mass

of

thc

whoc

har.

If

thu

load

bc

at

Jst])esa)nGnsa.iL-]igth('))i))g'ofthcb!U-mt~c (Compte 1U7.) prineipte jnay inertin hc appticd of n. ~ttifoi-in

to estimatc

T!)c duc

samc

t!)C

to ti)c rotatory rod. Wc havo <md what additton to m;)kc to on!yto thGkineticcncrg'y, sup])osing tha.t tho bur vibrtes t.o thc samu !:tw as wou]d oblai)~ accordin~ were Uierc no ioc'rtia. rotatory Lctu.s far cxmnpic, take, thu case uf a L:u- c!a)npc() frcc at a.nd assume tftat th vibration is of th type, .V = !< cus~ whcre
f'no-gy

at Oaud

M is one of thc func-tions

cftttc rotation is

invosti~atud

in 170.

Thu

].i))(-tic

18C.]

CORRECTION

FOR

KOTATORY

INERTIA.

23U

Tothismustbca.ddt.id

ti0 that

tlie

lunctic

encrgy

Is mcrcascd

in thc

ratio

Thc anee for

atto'ed rotatury ofthe!

frcqnency inertie prcce~ing. 7'= ~=

Lcnrs
n mlio

to

thi~t

calcnh~tcd
is

without root

allowof th

'\v!uch

thc

square

rcctprociti

Thus ?~/c' 1-(~,+~(1). cosh ?~ == suc M, sitilt ill thc furjn

eus ?~ 1 ces a /7r siu a' 1

,?</

M~\

By use wu m~y

of thc

retat.ions K'

?~==cun <'7r.t:).)t7~,

cxprcus

A\'L<j~ .-<;=
sin M + ces

;t.

eus

<7r

if wc substitntc

fut' ?~ front ~=~(2t-l)7r-(-l)'a.

In thc
nunutcs,

c~su of th K==172C',

~ra.YCSt wlicncu

tune,

ot='3()43,

or, in dL'grccs

and

Thus

which

~ivcs tlic gt'avcst tonc. WtK'u andtheti thc

corrcctitn)

fur

rotatory

incrtin.

in

tlic

case of thc

ordcr

of 'u=l

th

tone

is modoratu,

a is

vcry

small,

sc'nsibly, n r=l-fl
/w\?)~

atld

~-) in importance witi)

(3), thc

tl~at thc correction shcwing order of tlic component. In a.ll ordina.ry bars K Is on its square ma.y be ncgluctcd

incrcases

and thc tcnn verysma.!), sensible error. wit))out

dcpcnding

240

LATERAL

VIBRATIONS

0F

BARS.

[187.

187. from

Wben

thc

rigidity

and

to point point it, aJong gnera.! be expresse) tumiyticaUy, tlie method.s oi'St.ur<n gated by and

of a bar are variaDc dunsity t)ie nonna.1 functions cunn~t in but tticir nature tnay be invcstiLiouviHe oxpituned in 14.2. at any

If, as in 1G2, 7~ <tcnotc thc vanahie flexunU rigidity of the and the mass of the clment, point bar, pM~ whosc is wc nnd as tlie gcncral dUTo'entiiJ quation

Ien"-th

tho

effets

f'f

rotat'ry

c< cos! nonnat

wo ot)tain fonctions

ixo-tiubcmg as the quation

onuttcd. to (]ctL')-i))C

If

wcnssumcthc i'orm

t]):)t of' thu

in whieh induite Let ends, normal inturnal series

us

is IImited ofdcnnitc

by th

termina!

conditions )~,

to

bc one

of an

quantitles that of

suppose,

for cxamp!c, termina! values

thc and

har

is cliunpc'd v:mi.st). vatnc cntirdy

function

at TI)c

both first no sidc

onc

so tliat

th

~.<; lowest lies

nonnid

for which Las its function, root, so that t)tc vibration-curvc T)nj function bas t.second has

Jms on nnc
bas

of the eqnilibrinm-po.sition. intcrna.1 root, thu t))ird t!ie )' function gcncra!)y,

two

interna) roots.

roots,

an'),

1 internat

nor)n:d fnncti~)).s arc conju~a.tc, Any two dincrent wiH vanish when say, their product mu)tip)icd by over tho Icngt,h of thc bar. Intc'grated Let th fonu us uxn.)nine th
nurnber

tliat ~Mt7~,

is to and

of

rf'ts

uf

a funetion

/'(.)

/M
compoundcd function of

=~M
of a Hnite

+
number of

M-tnormal

+~

(.).(3),

of \\hich functions, tho lowest ordor is ':<(.) and that of highest ordct- is of internai roots of/(~) (-<"). If tl'c numbcr so that thcro be arc ~+4 roots in all, thc dcrived functiou (.?') cannot hve )css + 1 Internai roots besides two roots at thu extremitles, than and thc second derived fonction c-annot hve rots Icssthan~+2

187.]
No roots and least another internal

ROOTS

OF

COMPOUND

FUNCTIONS.

24L

can bc lost double roots.

whcn

the

latter

din'ercntiation Hcncc by

is multiplicd by 7~, with respect to x will ]cave at that (2) and (3) wc conclude
M +

function

M,

<

M.

(4<)

as /(.). Since (4) is a function of the same form as/(.), thc same argument in~ybe ropcated, a.nd{), a series of functions of whic)) lias at least obtaincd, every mcmber aa many roots as/(~) lias. When the operation by wliieh (~) was derivcd ft-otn (3) bas hccn rcpcatcd is ofteo, a function su<Rcient)y arrived at whose fo'm differs as !itt!c as wc picasc from that of thu normal function of highest order component ?<(?'); and we concannot have more than~-l Intcrual In cludothat/(;c) roots. likc manncr wc may provo cannot hve less than w-1 1 t);at/(.r) internf).! roots. of this thcorcm to
an

bas at least

as many

roots

The application
ofexpa.nfiinga.narbitraryfanctionin

deMonstrat
infinit

th possibility series of normal

functious

would

procecd

cxact)y

as in 14'2.

188. is subject

When.

th bar, whose

latral

vibrations

thc potential tension, of two parts, the nrst (ieputtding on the figuration stinness by which th bonding is directly and the second opposcd, on th reaction th extension, which is a neeessary accomagainst of tho bending, w])en th ends arc nodcs. Th second paniment to the potential the partissimUa-r energy of a dcnectcd string; first is of of thc hithcrto same in nature this as Chapter, tension. that with wtuch it is not \vc have entirely becn indeoccupied pendent thongh

to longitudinal is composed

are to bc considered, energy of any con-

th permanent th extension from

Consider whose distance

of a filament

of th

bar

of section

f~u,

on th plane of vibration projected which were normal is Since th sections, to the axis originally, remain normal th bending, th length of th niament during bears to th corresponding of th axis the ratio Tt* + lment JT, 7~ being tbe radius of curvature. in th ratio q :y-~y, reckoning 7'j dnote the whole tension Hence R. the actua! tension on Now from to thc which filament th thc axis the is itself bar is is ex tend cd state, subjected. if unstretchcd

the axis

-~+~(7'+~)~M. J (j

from
section

whieh

we find

for

thc

moment

of t))c couple

acting

across

tho

and

for thc

who)e

potcntud

cno'gy

due

to stithtcss

an expression
substitution

din'cring ofy+7'fory. is ttic tension loi~'th,

from

that

previousiy

uscd

( 1C2) t'y tho

Sincc twicc its Y'would T!te dnring ofthc whcn to th

illllolll'It amount a.

natnmt

to strc-tch a har of unit arca to ruquircd it is cvidunt th~t m most pra.cticat cases
iti comparison with

bc nc~tigibic

expression (1) dnotes thc strai~htcning'oftLe axis W(.'['o prc'scrvcd strctchcd L~r or strin~ position, thc

thc

work

th~t

\vou!J

bar, if th luugth t])(; proccss. But constant dut'ing to p:LSS frotn iL disp!a.ccd is attowcd uf t)tc axis the thu is dccrcascd. corrospol.lding corrcspouding Tho gain

ne ~iocd of c:).ch dment

nutura!

to)~t])

of th tho dco'casc clocrcasc

Is << and ( .") -cl.r, is :~f j\~t< G?fY : T( ~(~. d.c. cl,c

cfworki.s

Thus r=< T)ie ment Yariation

(~

.(~)' of the first

&+

-~(~,)'(a). dispiacc-

is givon

in 1G2.

part duc to n. hypothc'ticfd For thc second part, wc hve

icf/7 ~8 J V~ In limits.

~c=

f~Sy j that wc

=
hve

(~y~l ( to

~y~ J
consider,

f~-Vc r.. f<~

/o\ (3). at thc

aH <hc cases Thcgcncra)

~y vnnishcs

diircrentia)

quation

iHa.cconiingty

or, if ~Yc' put

-t- T =

~'=

~),

~+~ rlx <:t~ .vcl.~c 'fc/</ 1. r!t .t. Fora. rcfcrrcd more to th dctailcd writings investigation ofCtebsch'

p. 0. cl.c cAc .U clt fM..l..t.~

equation n.nd Don~i)i.
Leipxig,

(4). readcris

of this

tlie

~'()r<t'~t'r7'<(ts<)'ct<</M<fr7~[ir/)<'r.

16G2.

189.]
18'). tite If thc ends

PERMANENT TENSION.
of thc rod, or wire, bc chnnpcd, = 0,

24:3
!ind

tc)-)i)!)i:L) conditions fu-c saLisfied. ]f t])c nature of Lhe support be such that, wlutc th cxtrutnity is coii.stnuuud to he a, node, tiio'o is no conp!c
on! scnt nmst thc Le c:i.su

itctmg
strfught. ofa.

on thc b:u-,
T)tis string strctchcd suppo.sition

must

vanish,
i.s u.sua.Hy a.s

thilt

i.s to say,
to rcprc-

tho

t~(cn i)i

ovcr

hritt~cs,

manytxu.sic:).!

i)Lst)'nmcnts; must pa.rt~kc Le altogcther n'in

to cos

but ofthc

it

is cvidoit

that

titc

!ui(l that vibration, n ]n:).ttcr of Ijiditfcruncc. ancrai

gct

thc pfu't beyotid thcrcforc its lo~tit

bridge cannot

thc
wc

dif1b)'c:nti;t.l

cqu.Ltiou

wc tit-ke~pi-oportional

whi<hiscvident)ysatiH(n'dhy

if yat)dy" for??,

bc

dcto'mif~d. suit:d)ty vimisha.tthccxtt-etnitic.s. n "'=~'

~+~7!

T)tc

sanic

solution substitution

a)sn

makcs

By

wcnbtflin

-~+/W

(3),

which If as

dtermines
we soppose

thcfrcquc~oy. t))C Aviru innnitt'Iy in

thin,

OtaptcrVt., by startin~ ]f wc t)-e:tt ns a vcry perfcct m-xibUity. vah)c of?; is approximatc ,rr "= For rcpht.cc a. \vit'(.; of a)i([
circult'

wn.s i'ound

?r=~7r~ from th sma])

thc supposition qnantity,

same of thc

<'7r</f 1'

+'

-;7T~rc 2~ (rr-~}of radius r, in tc'rtns

scetion va)ups

~=

and

if w

f< hy thcir

of y, 7', an()

whic!) \vithin of th

gtvcs
brackets

t!)c

corruction
invoivcs ?',

t)) ri~idity'. 1. Since It appc'ars that t])C harmonie by thc stiitncss.

fur

expression
rclatinti

componcnt

tones

is (Ust~rbed

'Dont{in's.-frn)f.f'f~,Art.im.

]f!S

244

LATERAL

VIBRATIONS

OF

BARS.

[190.

190. ends

The

investigation

ofthe

correction

for sti~ncss

when

the

In consquence of is not so simple, ctanipcd In on)< to puss occnrs ncn.r tlie ends. of type which thc change connow undcr section to that from the cftsc of th preceding with the be introduced, must consti-:unt au ~hiitional sIdGration Die fu!tow!ng the pitch. is, in the eHcct of attti fm-ther raising ofthewirearc ma.in, thc If the becomes investigation rotatory of Scobcck incrtia be and Donkin. th differential quation

neglected,

where Th

a and

/3 are

fonctions

of ?t determmed nm<!e to

by (2).

tho four boundary sattsfy as therc are only three ratios, tca.d which, clisposable conditions, This may be put into th form to an equation a, ~3, connecting solution
must

now be

190.j
Thus the now far our dincrential ititroduce

PERMANENT

TENSION.

245

equations

are

rigorous, are

or ruther

as

on which thcy equation that tlie the supposition by tlie existence expression

founded; vibration considered Tttis being

as rigorous but we sha.11 is but th case,

aSected slightty t!te approximate

of rigidity. for y is

uearty. Th (G) proves introduction that H' 6~ < of thse ur .j values

is a

into
stna]]

th

second

of equations under th cir-

quftntlty

cumstn-nces a.nd thei'cforc tli:tt a'~ is a l:u'gc (~tfnitity. contempiatud, Siucc sinha~ are both I~i'gc, ('(~uation to cosha~, (5) rcduccs

According pitch by tlie tion wise follow rigidity only that is not

to this same

thc component tones are ail raised in quation smaU interval, and thcrcforo the harmonie relath It would probably be otherrigidity. f were reta.incd it does not therefore relation is botter in spite nf preserved are frec, but former case, It should quantity, enough

disturbed by if terms involving that when tbcre thc the absolute that b harmonie ends

is no

than when they ctamped additional disturbance in th of pitch is much (t or ~/(<y + l') \/7', is is to be correct, A: rnust. a and yct romain

are

tlie though be remarked and

altration

greater. a large be small small.

that, if our rcsuit to bear multiplication

by b

346 Th

LATERAL theoretica-1 rcsult

VIBRATIONS cmbodicd who

dmtuccd

0F

BARS.

[190. with Thu

experimoit by Seebeck, constant of stirfness \vas

oi't)io \'ibr~tions uf

found

in (8) ha.s been eompared a. s:d.infactory agruemcnt. frum observations of tite of thu \vii'C) wttcn

n smaU

piuco

rapidity one end was

(.'tutiipud 191.
cvcn whcn

lit tt. Vice. It lias hcon

sh)]pUHud

shewn
to thc more

ni t)ns
utrnost

c])apter
by t)tc

tliat
omission

thc theory
of

of bars,

uniniportant

quantifies,
fiuxibtc

is (tceu~dty The stnugs.

cumpticated

t!t:ut extrme

ruasun

of

thu

of .strings is to Le fcund in thc flet luu-monic type arc propagated of tho with a velecity. indcpL'nd(;)it wave Iun~t](, so tti~t an a.rbitnu'y wa.ve is aHowcd to travut Avithout But whcu wc pass from string's dcomposition. to b;u's, t!ic convibrations stmt stant longer
transmission

po'ftjct!y of thc snnpiicit.y titat \),v(.'s of tho

t.hat

of

iu in

tlle thc

(litlrential din'crGntifd as
a train

C( lliLtloll quation,

`l ~'1Z. -Ivix. ~t-/<=(), and

wavc.s

`l t])(j
dpend

= U is is no V(.d"city
on

cxprus.sihiu
of

a
of

veh'city,
harmonie

thcrufm-c
c:mnot

of
th

dif)'cr(;utial Indccd, if bc propagatcd

ontution ,'dom', but it bt.' admittud t])at a.t a)), titis

must t))e

vary with tiiu train uf harmonie is sufHck'nt

wa-vc lungth.
wavcs can

to by itscif must vary inycrscly a.s thc wavc tcngt)). provc that thu velucity Thu samo front thu soJution to titing may bc scen npj)]ieab!c C) wavcs in onc propagatcd direction, vlx.=cos". (H~),
which

considuration

satisfies

thc

diH'urcuti.d

C(p(ation

if

u.s suppose but. uf amp)itudL's, directujn. Tiius

Let

that diftbruut

titcrc

tu'c

two

tminn

of

wavcs m

w~vc

]c))gt)).s,

trnv'L'HI))~

equa.1 t.hc samc

of

If

r~

bc

.smn.1),

we

ha.ve

train

of

wavc-s,

Avitose

nmpti-

ouc: point to anothur IjctwGOl thc vatucs 0 amt 2, ft)!')ning- a so'ic.s of group.s S(-)):).r:).ted from onc aufjther by frcu ironi In t)tc case of u. ]'egiot]s cojnparativ-cly distm'baucc. of a co]um!i v:n-ics as T.and t!)cn thc gt'oups move stringor ofair, s!(jw)y

tu()c

vancs

from

~91.]

0F TWO TRAINS 0F WAVES. RESULTANT

n.s thc velocity It is ot.ttcrwise
compone~t trains, :t.nd

347
t!ierc

forw:u'd with titc same is no change of type.

th t.ransvcrse!y, vdoctty vibrating ot'thc wave Icogth. Titc position at ti)nu t of thc is at thu origiii givoi. hy grnup which was initiatty

whcn, as in tiic case ot' a bar of' propagation is a fmictton middia of t)t0

In thc gt'oup.s H)2. tion time on

cn.SL- )!.== 1, an<t accordingly j'rcscnt is <t'ce that ot' <,)nj compoount w.~ves*.l, On account tho wave of tho (tt.'ppn<)cncc' tin.! cutHution

t)in vc!ocity

oftiio

subsc'j'tcnt will h:tve purtiott, cort'cs~ondmg i)ivcstig:).tion It sfitisfv

(tistnrbancc f'ontuu'tt tu a, lunitcd initm.t tho n<jnc of t.hc simplicity wttich chanicteri.sGS ior a .sLriug'; bt'.t ncvL'rLhutcHS Fouricr's pt'obtt.'m umy property qncstton to dutcrmmc a. function <)ttd :t.p):).cu hci'c. of :nid t, so us to

Icngth, to an

of thc! vclocity of :ui infhnto

of propaga.bar at :u)y

ofthis

I.s rcquircd

and
A

!U:~kc
solution

initiaity
of (1) is

(.),

~='

(~').

~/=cos~

cos~(.<x).

whcrc

and

arc

constants,

irom

'\vluch

we conclude

t))at

In

tho

c<in'csponJh)f} Yarit~

pr')1))om

fur as

wfivcs tho

0)i square

th

surfaco root tLcrefuro nf

\'dot;it.y of prf)p!ts'~t't"~ that Tho vetocity M=A. tbe component trains.

dh'(;Kt)y tt of ( group

of Jcfp water, tho W!t.vo Icut;) f~<; of thttt

tho su of

such

Wftvcfi

is

248 is n.)so

LATERAL a sohttion, where

VIBRATIONS j~('x) is an

0F

BARS. function

[193. of a. If

arbitrary

!iowweput<=(),

which Founer's lience

shcws

that

~(a) intgre

must

be takcn thcorcm

to be 27r ~j,=~(A').

(a),

for

then

by y=0;

double

Murccvcr,

By

Stokcs'

t!morc!n

t])e suppty (liU'ercutml

i-(;m:umng quation

( f)5), p:u-t of thc

whilc

we mny now iudependently, which I)ns to Ha.tisfy tlie sulution, = Ib makcs initi:d)y 0, = (.); it is

or

Thc

Hnal result (4.).

is obtained

by a.dding

thc

right-I~ud

members

of (3) aud Jn mctuis

(3) thc intgration of thc formula.

with

respect

to

q may

bc

c~ected

by

which formula

may

Le proved

as follows.

Ji' in thc wcM-known

intgral

Now only th

that~=<'=< suppose ren.1 p:).rt of tho equation.

whcre TI)us

!:=VI-l,

and

rctain

193.]
whencc

FOURIER'S

SOLUTION.

249

from

-which

(5)

foUows

ecma.tiou

(3) may

by bc wi'tttc]i

a. simple

ct~nge

of

va.nab!c.

Thus

CHAPTMR

IX.

ViiHATtONS

0F

MHMtiItANES.

is a pcrfcct]y f)cxib!c and inJnmina ofsotid nnitctythin nnifoDn )nattcr,of materiat and thickwhicb is strctcbcd in :dt directions ncss, so grcat by a tension as to rem~in scusibly unidtcre.t th vibrations during at)d di.spfaccmonts If fui imagioary Une bc drawti eontcmpjated. across t!ie membrane in any direction, eiemunt t))omut)):d action betwccn thc two portions

U)3.

Tm-: tlicorctica!

monbranc

separatcd byan thc dment and bti l' gestion it i.s a quantity Ti)c

uf U !inc

pcrpcndicutar
ditncn.siun

~.9, 7', mnybo ofont:

is proportionn! to thc len'rth of to its direction. 1. If t)ic for~c in caifcd tbc <o<uM of ~e ??te~6~Hein tnas.s and2 in time. with tins

is tlie snbjoct vibrations of mcmbratic.s of dirbrcnt arc nxcd. shapc.s, whosc boumhu-iu.s Otbcr ind~cd questions rnay bc but thcy arc of proposcd, Htt!(j intcrcst; compat-ativuty and, t))e tuutttod.s for niorcovur, prop~'r thcm wi)[ be 'suff~ sulvin~ in otticr parts of this work. cicnHy iitustratd Wc may titcrofor procuud at unce to the con.sidcration ofa membrane strctchcd ovcr thc arca inc!)tdcd witbin a nxcd, closed, ptanc bound:u-y. 10~. dnote monbranc. acting upon Taking th smail as t)iat of a'y, let M bonndary tbcret'roni of any point disp!aco;ncnt 7~ of tho Round takc fi sma)t an~ amt considothc forces it parattcl T)~ to z. of the tension roso)ved part is phinc t)tc of tbc

principat probfon uf tbc trau.svo-sc investigation

in conncction

cxprcsscdby m f~
~'j~ wltero cfomjnt balanccd (~ dcnotc.s "m~ r/~ bound~y t/ of ) and tf/t tm nn ot thc normal to thc cnrvc drawn out\ar<).s. This is by the reaction accctcration against mcasnred by ~v an ch-mont of tbc

194.]

EQUATION

OF

MOTION.

251

p buin"'
dcnotmg

in mass 2 hi uf onc dimension a.nd a symbot Nuw by Grcen's tlie supui'Hci:d theorem, density.

length if

;S' ukimatufy, f).ud thus th cf~un.tiott of motion is

.d).

~)

Thc Thc expression thc totisiou

condition diU'orential for thc Ly th

te bu s:).tisficd quation putcntin.! supediciid

at tlio bomidm'y ma.y a!so winch

bc

is of course
from

w=

0. thc

invcsti~atcd

cncrgy,

strctclimg.

by muttiplying a.rca. is T)ic :dtcred

is fouud

from

which

8~

is casily ?~ /]=c'

ibund thun is

Ly

au

intL'gratIou

by parts. of tt.vc!ocity,aud

If wc writc tlie diH'cruutial

c is of t)ic nature

con~tion

of th memthat tho boundary now suppose formcd by Llic cnordinatc axus and thc linc.s is thu rcchuig!~ brane for ovcry point withhi tlic arc:). (:}) 104 is satisiicd, te = n, y = 'w=(). fmd fur cvory point ou tnc boundary !!)!'). We sha!l A particuttu' It~tegral is cvidcntiy

l 11 C l'C where

?l-

C-7T'

,7)~ /))Lz 2

?~\
CI) +~(~')l21

and M~ and H arc intcgcrs Thus dcrivcd.

w=~ w=S M-i )<=~o S M=t siu ??;7TT'

n.nd

from this

thc gnerai

solution

ina.y bc

);'77' {:Imn sln-(~~cos~<+7?~sin~}. COSI)~ -j- En," Sll1 2)t} (3).

252

VIBRATIONS

0F

MEMBRANES.

F 19 5.

That

this

result tliat

by shcwu~ circumstanccs. WI~tevcr presscd

is really gnerai a posteriori, may be proved it n~y be ad~p~.d to initial express arbitrary of thc of

fiiiietioii

for all vaines

co-ordinatos bctwccn thc limits

may hc, it can bs ex0 and by th sries

where whatever

th

coemdents function

Y.,

be expanded

of~nnyoac betwecu 0 and

Indcpeudont ofthe coc~cicnts & iu t)ie series

&c. are

of YmayLe,

Again it can

C, &c. arc constats. From this function of x and y can bc expressed within angle by th double series

where

we th

conclude limits

that of the

any rect-

and

thcrcforc initial

that

tbe

arbitrary

values

expression of w and

forain In fact

(3) eau

beadapted

to

.(4.).

Thc

dmmctcr

of tlie normal

functions

of a given

rcctang!c,

is easily undcrstood. and If and n be both thc same sign over th whole of th rectanclc at th vamshing but edge in any other only case there are nodal lines to the axM of running parallel coordinates. Th numberofthc nodal lines paraHetto is n -1, their equations being

as depending on w retains un.ty,

195.]
In n)'f. th sa-me

RECTANGULAR th

BOUNDARY. noda.1 lme3

253

way

quations

of

th

pMa.Hcl

to

1 in number. w being into ??~ equal parts, in repeated. 106. 1s ~q Th expression

divides system ea.ch of which th numcnca.1 The nodal

th

rectangle value of w is

for w in

terms

of th

normal

functions

whcrc

6,

&c. are for

the Fin

normal terms of

coordinatcs. We

We hve

proceed

to

fonu

the expression

over th area of tlie rectangle these expressions integrating and we find coordinates of th normal disappear, the products In

the

summation The

being

extcnded for th

to <d! intgre cncrgy

vahtes is proved

of in

w and th

?!. sMne

expression wr,v to be

kinetic

if Zf~cf~

dnote

the

tmnsvcrse

force

acting

on the

element

~.t.-<

254 Let

VIBRATION

0F

MEMBRANES.

[1()Q.

ns suppose that, tl.e i.iitial condition is one of resb undcr t)tc opration uf a consent, force .s.ie], as nmy Le to supposed ansu from gascons At thc pressure. tin.c <=0, tlic i.nj~d force is rc.novcd, ~nd thc mo.nbmuc Jeft to itsc)f. thc IniMativ is cquation ofcquiiibrium

i"C('njunctinnwit,h(~. to cxpn.s.s v.-Lh.c ,n (..), or ,n t.).is J'iLpgr;d.s)gn.

Thus

In

on]cr

case

si,nr]y

to

ren.uvc

f,.o,n

undcr

thc

Thi.si.s If tlie solution is

an

cxamph.of

(.S), {;)()! p,i.s)y at. n.sL in its po.s.Uo. tlie

,nen~r.n.. set

ofcn.i),-

4
=~

W7TX

M~S/Y .sin ~< .(~).

197.]
197. ascribing The

CASES

OF

EQUAL
thc na-tural

rERIODS.

255

frpqucncy diiTerent intgral

of

vibra-tions in tho

is fouud expression

by

values

to M and

For Hido mode, more tu th

n, givcn. mode of ttie rectangle when

of vibration is incrca-scd. additions th

iurni

thc

w=], ~=~ n.nd whcn effective; longer

sidc {u'c are a-tmost

faUs whcn cither pitch In th case of thc gravest to thc shorter Hidc iirc titc
very

is

clo!)~ttcd,

additions

wlthout

c~uct. uo two ci values pairs cach fuodamcntal

WitCH of w and ]t)cdc ft" a)Kl may hve

a~d ); can arc t)tu In gi\'c

inconnnensur:ddc,

t))C sa-mc
ils own

frcqncncy,
ch!Li'actc)'istic

and more

of vihratiuu

bas

coonnmisurabtC) samc pcriodic thc suc)) casus

two

or

Uni whctt pcriod. fut)d:uncnt:d modes tlien

ils

proportions, charaetcr. not thc of

w)fi)e

ti)nc, and may motion sti)) rctains thc Lhc sp~'incxtion Thc type.

cocxist
simpte

any harmonie docs of

in

of fuM thc thc

thc aid

co)np)ctc)y prohion now

detenninc

pc'riod cnnsidcration of thc

Lut numhbr.s; to considcr work

it~c]f n'nnircs prcscntin~ for it will bc sufUcicnt a few of thc

thc

membrane

in Ricmann's If f; =

Th is square. lectures on partial

cases, sunpk'r rcadcr will find diUbrential

purposes which arisc fnHci'

thcory of this whcn

information

quations.

Th which

lowest givcs only

tone onc

]S foun<t
funda)ncn<d

by putting )nodc

n~ and

)z cqual

to unity,

Next the are be other

suppose that to unit.y. '\v!fosc

one In

of the this

numbo's

obtfuncd, synctironous

po'iod.s thc in phase,

way two arc th s:unc. wt)ole motion

?)!, ?; is cqu:~ to 2, and (tistinct types of vibra-tion If Utc twr) vibrations by is exprc'sscd

so

that, liarmomc

although motion,

every thc type

vibrtes synchronousiy part is to somc cxtcnt of vibration

with arbitrary.

35G Four particular

VIBRATIONS cases may

0F

MEMBRANES. noted. First,

f'197. if 2? =0,

be especially

which Sinailarly edges. equal.

mdicn.tcs if C'=0, Ncxt, Thon

a vibration we have

with a node

one

node

howevcr, suppose w is proportional to

parallel that C' fuid

along th line to th other D are

a?==~.

pair of nitc and

which

mn,y Le put

into

th

furm

This

expression

vauishcs,

whcu

or ngfun,

\vhnn

The

quations give the edges, assumed to be nodal while the third gives one diagona.1 uf th square. In th forn'th th cdgca lines, The figures represent case, of ttte when C= together cases. D,

first

two

which ~+a*=

wcre

originaHy

a, representing for th

we obtain with the

nodal ~=.r.

square t!]e four

diagonal

c+~=o.

For is corvcd,

other but

relative is always

values

of 6' and

7) thc

interior by

nodal

Iine

a.nalytica)]y

expressed

and may he casily mic cosines.

constructed

with

th

help

of~

table

oflogfu'ith-

197.J
Thc Tiie their next case values

CASES

OF

EQUAL

PERIODS.

257

in

ordcr n

of M~ and

-\v]nlc mtcrchangc, of vibration. ft-equcncy accordIn~Iv

= 2, = 2. no altration is caused being equal, by no ottter pair of values givcs the samc to bo considered Th is oniy type of pitch occurs w!icn

whose

nodcs,

Jetct'mincd

by th

equation

arc (in addition

to the cdgcs)

th

straight

lines

T~)c next

case

winch

we shaH

consider

is obtained Wc

ta w, n th values f~. M) = Usin ( The nodes

3, 1, and 37ra; a sin

1, 3 successively.

by ascribm"have

Tra; Try 3~~ sin + D sin cos M<. a o o J

arc given

by

or, if we reject

the

first

two

,{a.ctoys, which

con'cspond

to th

cdges,

which

represent R.

th

two diagonals. 17

258 Last)y, if C'=

VIBRATIONS tlie quation

0F

MEMBRANES. of thc nodc is

[107.

whcn and similarty Jn ca~c (4-) wlicn a: = a, y = ft, or TL'us oue ha!f of Ctich of tlic lines julning y = a, ? = , or by th curve. tlie xuddie points of opposite cd~cs is intcrccpted !'n,tio to one another th~t in wha-tever It should bc noticcd th Le t~kcn, thc funr points of mtcrscctio!i If the D=0. C'=0, cases, Mid D may with correspon~ing pounded of Fig. shaded compnrtmcuts nodfd of thc vibrations phfmcs, (3.')) tlio it eurvc always nod~I lines of thse is vident directions passes of tlie cases tha.t through Urst two 'bc in comthc

of disph~cment

nocM curvc no pM-t of the thcrcfore und that s~nc, thc of amplitudes, thc ratio whn.tevcr ia to bc found thcrc; When thc utish~dcd Jnust bc drawn curvc portions. tlu-ough tlie nodal curvc will ~]-G opposcd, hand the phases on the othcr thc shadcd portions. through p:uis Gxelusivcly n.rc thc tlie nodcs w =3, ?t=3, tu th ed~cs shown in Fig. (3G). When M-e thc straight lines par:illct

197.]
Thc iMt

EFFHC'T 0F SLIGIIT IRREGDLARITIES. 259

ca~c which we shd! consider is obtaincd by putthi~

or, if th factors

con'esponding

to th edgcs

be rejected, -1 -0.(0). Systems paraUel of th to titc

o 4cof:l -1 (4 M cos-.+Deos-o c(4co~l)cos~+Deos~~c~l)=0.(0) <x\ 4C08 If C or D vania!), wc feU! back on tl)e eomponent edgcs. vibrations, If (7=~, our

ce a

nodai

of straight lincs consisting quation may bc written

of whieh the second If diagonal'.

the

nrst

factor

a hyperboHc

rcpresputs curve. the same

the

diagonal

~-)-~=~ to t!)e

a,nd

(7=-7),

wc obtain

figure

re]ati.vc!y

othcr

modes of a. sqaa.re membrane, pitch of tlie natural which is nearly, but not quite aniform, may be nn'estigatcd by t he geucra] method of 90. in. thc first place tha.t w a.nd ? M'c suppose equal. In. this case, when thc pitch of a umform membrane is givcn, the mode of its vibration is comp!etc!y determiued. If we now conecive a variation of dcnsity to eusue, the natural of type vibration is in gnera! but thc period modincd, may be calcutated without aHowanco for th change of type, approximatcly Wc have We will

~98.

Titc

of which ifwoecos~ we have p~

th

second

terni

ifl the th

increment v~lue cf~

of

T due

n.nd P dnote 9. T) 1 s ,,en=i 1=~

previousiy

to 8p. Hence to variation,

4 4ff"8p p (1 o. rn n PO ~s'

~H7T.T .B~~V, 7 m~r_r. ~in2 ~~z~r,y ~,r,rly. ~a (1 ~y.(l). a

(1)~

'It()n)~,J')).<<tfrt'c~.<<'<<{?,p.l29.

17--2

2GO

VIBRATIONS

0F

MEMBRANES.

[198.

For the

exemple, square,

if thcrc

bc a small

load

Jtf attached 1 r

to thc middie

of

in which th duces

sin~ ~~Tr vanishes, la th former bc odd. unloaded no result. however, to subject from for th M and the membrane,

if case and

be cvcn, thc centre thus th

and

is cqual to unity, is on th nodal line of th load

if of pro-

addition

When, maining liarity raturai fied

n arc

uncqnal,

the problem,

same

different type

anything unloaded

principles, gencral we have hithcrto membrane

rethough a pccnpresents met with. Ttie to a speciintroduction In lment.

period is now to some extent removc the indeterminate load will in gnral ofthe of th undison tlie assumption th period to calculate attempting of tho undiswill arise how the selection turbed type, the question there are an indefinite that is to be made, turbed secing type of th membrane in th uniform condition which give number, is that those types must be chosen Th answer identical periods. under little from th actual types assumed Innnitely of th load, and such a type will bo known th operation by th calculated from it a maximum of its making th period criterion which differ or minimum. a simple to th attached that two we wish equal As example, membrane what of tlie let us suppose that a small load

corresponding bnt the arbitrary;

Jt~ is

at a. point

to know

periods

periods unloaded membrane, or

and lying on the line x = for t!ic are to be substituted found by making

= 1, M == 2, It is clear nodes that the normal types in th first

?~ = 2, M= 1. to be

are represented in th first case th incroase it can bc; is the least that is the grtes If /3 be th possible. is altered in the ratio energy

whose arc those chosen, In tlie two ca~es of Fig. (32). due to tbe load is zero, which period case the increase and in th second ordinate of Jf, the kinetic

198.] whilo e Tito toucs ratio

SOLUTIONS

APPLICABLE ?)'=P'-

TO A TRIANGLE. 2

261

eLaractcristic

of thc

interv~l is thus

betweoi approxnna.tcty

t!)c

two

uaturnl

of th loadcd

membrane

If

= ~<, ncither As another

pcriod

is aHected

by the

load.

thc case, where th values of w and example, are 3 and 1, considered in 197, may Le referred a. !oad to. With in the middie, to bc seleetcd are those ttie two normal types to thc last two cases of FIg. (3't), in th former corresponding of winch The the load has no efTect on tlie the period. vibration of a square memand heavy load is more dIiHcuIt, But it may be worth while tu the actual than period is greater a hypotlictica.1 type, winch dinars

of determhung probleiii brane winch carries a relativcly we shall not attempt its solution. rccali to auy ttiat froui tlie 199. good dcal thc fact metuory can hc calculatcd actual The more one. preceding than was tlicory at first that from

of

square iutcudcd.

membranes Wheuevcr

ine!udcs in a vibrat-

:).

remam at rest, t!iey may be supposcd to ing systom certain parts be absohitelynxed, and \ve thus obtain solutions ofothcr questions than t!)osc origmaUy For example, in th present case, proposed. 'whGrcvcr of th sqnaro is nodal, we obtain a sohttioti a diagonal is an isoscelcs boundary to Morcovcr, any mode of vibration possible right-angled triangle. tho triaugle to sotno natnnd mode of tlie square, as corresponds tlie vibratwo triangles may ho scen by supposing put togcther, apphcabte tions each stances witttont cludcd Thc puttiug being other it at points equal and opposite lu thc common hypothcnu.se. is evident that th which are Undor of images thcsc circumat rest is iuth to a membrane whoso fixed

wou!d remain bypothenuse aud tl~crcfbrc tlie vibration in question constraint, those of wttich a complte among square is capable. frequency ?~ == I, n= of thc 2 in tone gravcst the formula of tlie triangle

1s found

by

and

is thercforc

coud '1

r/~ to 2ft

2G2

VIBRATIONS

0F

MEMBRANES.

[199.

Thc

next

tone

occurs,

whcn

M =3,

?: = 1.

lu

this

case

as might into two, triangle For ary to Lamd's of thc which whosc 200. th of may first thc

also

bc

seen

FIg. (37), in the ratio

that th triangle by uoticing whose sidca arc Icss than those \/2 1. vibrations of a membrane

dividcs of th

itself whoc

tho tlicory is in thc form

'Levons

of thc

whose

bound-

of an cquilatend sur l'lasticit.' /t, wlicrc frequeucy

tone is c gravest is thc same as thc diagonal is A.

is refcrrcd th reader triangle, It is provcd that th frcquency of th trianghi, A is tlie hcight of tlie gravest tone of a square

thc Whcn step towards

gnral be effected

is circular, of thc membrane boundary is thc expression a solution of the probicm This in co-ordinates. diHcrcntiaI polar cquation but it is simpler to form the polar analytically fixcJ

thc forces whicli act on th potar de novo by considering As in 194- the force of restitution of arca ?' dO t~ acLing is on a small arca of tho membrane cquation etemcnt

and

thus,

if TI

= c" as before,

tlie equa-tiou

of motion

is

subsidiary w=0,whcn?'=f/. In order to ha.vc nn\v 't'hus, /)

The

condition

to bo satisncd

at the

bouadary

is that

invcstigatc that to assume

cos(~<e), c = /< the rhf'crcntia!

if ~cc

we vibrations component fonction of thc time. is n harmonie of brcvity we writu and for thc sakc th normal cquation appcars in the form

200.]

l'OLAR

CO-ORDINATES.

2G3

In which Now it eau

is thc whatevcr

ruciproca.1

of a liucar nature series

quantity. of as & functiou of ?' and

be cxpMiJed

ma.y bc tlie lu Fonrier's

M = w. + in which uf snbstitutiug

cos (~ + al) + M~ cos 2 (~ + a.~) +.(3), but not of (:;) In (2) may be written fuuctions of The result

&c. arc froni

thc

summation

this multiply to betwuen vanish function

mtcgra.1 by ces M (~+ aj, and quation 0 and 27r, wc sce th limits and we thus obtain to

cxtcnding

to

all

values

of

?:.

If

wc

witli respect integrate must thttt each term dotermmG as a

separately, of r

in

which

it

is

a mattcr

of

indirfcrcnce in

whcther or not.

the

factor

cos n (~ + a,,) bc supposcd Th solution of

to be includcd

functions of r, two distinct (4) involvca But one of thse constant. cach multiplied by an arbitrary and the corresponding Infi nite when )' vanishes, functions becomes the as not satisfying solution must be cxctuded prcparticular This point may conditions at th origin of co-ordinates. scnbed bc illustratcd (4) by ruduccs satisty grated tna.uner derived from to the simpicr equation by a roforeiice solution in question when the K and ?!. vanish, making does not at tlie to to=Iog?', origin which, however, = 0, as may a small bc circle scen with of th from the the value of In for centre intelike our of

\7~ round the

prsent tlie membrane Th function i)i from several thc

comptctc since purpose,

Intgral it covers

origin for centre. (4) is too gencral case in which th force. (4), is the bc

is subjected function of

to an exteriial )', which dcnoted by

othcr ofthc

satisfies

Bessel's cxpressed

border, Th ways.

difrerential

asccnding is quation)

(~?-), and may sories (obtained

nnmcdiately

264

VIBRATIONS

OF

MEMBRANES.

[200.

which t!)at

is Pessel's

J,, and less than umty. Ttie

ongiua! its differchtia!

form.

From

this with

expression respect

it is vident to z are aiwaya

coe~eicnts

sries (.5), though for all aseending InHnitc, ia convergent values of~ aud z; but, -\vhen is grca,t, the couvergcncc does not as a basis long time, and then th sries bccomes useless Lcgin forn. for nuincrical calculation. In such cases anot)ter series procecding l)y desconding This sries is powcrs
of may Le suLstituted

with

ttdvantagc.

if 2~ bc cqual to an odd Intcger, but It terminates, otherwise, runs on to innnity, and becomes ultimately Neverthelcaa divergent. wlten z is grent, th convergent in ca~culapart may be employed for it can be proved that th smn of auy nuinber of term~ tion differs from the true value of thc function by less than th last tnn inctuded. Wc sba,U ha.ve occasion with later, in connection to consider th drivation ofthis descending series. problem, As Besscl'sfunctiohs are of considrable in thcoretiimportance cal acoustics, I have thougbt it advisahie to give a table for thc functions extracted from and LommcI's' due work, J,, and Lommd, $<;<(~'< &cr clic /?M~'t-c';<) FtOtc~fn. Leipzig;, 1868. anothcr

it

200;]

DESSEL'S

FUNCTIO~S.

2G5

to Hansen. originatly thc rela-ticii

Thc

functions = J~.

J,

and

J,

arc

conncctcd

by

~)

~(~

~(=)

~.(~

') ~(~_ 9-0 f; 9-1 9-2 9-3 9-4 9-5 9-G 9-7 9-8 9-9 10-0 10-1 10-2 10-3 10-4 10-5 10-6 10-7 10-8 10-9 11-0 11-1 11-3 11-3 11-4 11-5 11-C 11-7 11-8 11-9 12-0 12-1 12-2 12-3 12-4 12-5 12-6 12-7 12-8 12-9 13-0 13-1 13-3 13-3 13-4 -0903 -1142 ~) -1367 -1577 -17G8 -1939 -2000 -2218 -2323 -2403 -2459 -2490 -24% -2477 7 -2434 4 -23GG 6 -2276 6 -2164 -2032 -1881 -1712 -1528 -1330 -1121 -0003 -OG77 -044G --0213 +.0020 -0250 0 -0477 -OC97 -0908 -1108 -129G 6 -146U -1626 -1766 -1887 -1988 -2069 -3129 -3167 7 -2183 -3177 7 -2453 -2324 -2174 -2004 -I81G -1G13 -1395 -116G -0928 8 -0684 -0435 +-0184 --OOGG -0313 -0555 -0789 -1013 -1224 '1422 -1604 -17~8 -1913 -2039 -3143 -3225 -2284 -2320 -2333 -2333 -2290 -2234 -3157 7 -3060 -1943 -1807 -1655 -1487 -1307 -1114 -0913 -0703 -0489 -0271 --00.~2 +-01G6

0.0 0-1 0-2 0-3 0-4 0.5 0-6 0-7 U-8 0-9 1-0 1-1 1.3 1.3 1.4 1.5 1.6 1-7 1.8 1.9 2-0 2-1 2-2 2-3 2-4 2-5 2-6 2.7 2-8 2-9 3-0 3-1 3-2 3-3 3-4 3-5 3-6 3-7 3-8 3.9 4-0 4-1 4.2 4-3 4-4

1.0000 -9975 .9900 -977C -9604 .93~5 -9120 -88)2 -84C3 -8075 r) -7~2 2 -71!)C -C7)l 1 -6~1 -MG9 9 -~118 -4554 -980 -MOO -~818 -2239' -!66C -1104 -0555 +-002;') --0484 -09G8 -1424 -1850 -2~43 -2601 -2921 -3202 -3443 -M4:) -3801 -3918 -3902 -402G -4018 -3973 '~) -3887 -37GG -3610 '3423

0.0000 -0499 -0095 -1483 -I960 -2423 -28G7 -3290 -3(!88 -4000 -4401 1 -4700 -4983 -5220 -541U -5579 -5C99 -5778 -5SI5 -5812 -57C7 -5C83 -5560 -5399 -5202 -4971 -4708 -4416 -4097 -3754 -3391 -3009 -2613 -3207 7 -1792 -1374 -0955 -0538 +.0128 -.0272 -0660 -1033 -1386 -1719 -2028

4-5 4.G 4-7 4-8 4-9 5-0 5-11 5-2 5-3 5-4 5-5 .6 6 5-7 C-8 5.9 G.O G-l G.2 6-3 6.4 6-5 6.6 6.7 6-8 6.9 7-0 7-1 7-2~), 7-3 7-4 7-5 7-6 7-7 7-8 7-9 8.0 8-1 8-2 8-3 8-4 8-5 8-6 8.7 8-8 8-9

-3205 -2i)Gl 1 9 -20!)7 .1776 -1443 -1103 -0758 -0412 ~? --OOG8 +.0270 0 -0599 -0!)17 .1220 -150G -1773 -2017 i -2238 8 -2433 3 -2601 -2740 -2851 -2931 .2981 -3001 -2991 1 -2951 -288~ -278<! .2663 -251G -2346 -2154 -1944 -1717 -1475 -1222 2 -0960 -0692 -0419 +-014G --0135 -0392 ~), -0653 3 -2404 4

.2311 6 -2985 -3147 -3276 (-) -3371 -3432 "~) -34GO -3153 -3414 -3343 -3241 -3110 -2951 -27G7 7 -2559 -2329 -2081 -1816 -1538 -1250 -0953 -0052 -0349 --0047 +-0252 -0543 -0826 -1096 -1352 -1592 -1813 -2014 -2192 -2346 -3476 6 -3.580 -3M7 7 -2708 8 -2731 1 -2728 -2G97 -2G41 -2559 -2791 1

266

VIBRATIONS

0F

MEMBRANES.

[201.

201. the

lu

accorda.nce

with

thc

notation

for

Bcsscl's may

functions thercfure bo

expression written a.]ul tlie boundary

for a nurin:),! ?(/=P~(/t)') condition

component cos~~+cf) requircs that

vibration

cos(~+e).(1),

~(~)=0.(2),
an -whose quation tli re fore of~). The complete roots givu for the admissible values of /c, am

expression

w is obta.ined

solutions embudicd in (1) wit)i all particular und M, and is ncccsstn'Hy to cove).' any initif), general enough circumstanccs that be We conclude tliat may imagiucd. an~ i'Huction of r and 0 may be exp:mdcd within tlic limita of thc circle ?' = a in the series

th( by combitung admissible values u:

~=S~~ For given arbitrary. Thc way. dtermination SInce tl energy overy by (2) intgral and for

(/er) (~ cos 7~+-~sm~).(3). -aluc of ? thcrc are a series cach of these tlic constants is effected to

of values <~ and in thc

of arc

of the of the

constants motion

usual

is cqua.1

and terms Tims, find

whcn

involve

expressed their squares,

by mca.ns ofthe it fullows tliat when (3) integra.tcd by

normal thc

co-ordinatcs

can

in (3) vnnisbes, if wc multiply

product over tlie

of any two area of tho and integratc-,

only of tlie circle. wc

~(~')cos~

-7r~[,7:.(~)r~(5), is dctermincd. Thc corrcsponding by w)iich formula, for -~r is obtaiMcd sin~ for cos?: A mctiiod of cvaluating hy writing thc lutcgral on the right will be givoi and prcscntly. SiucG cacii contam two terms, one varying as eos~~ and thc other a~ it is now vident how t)ic solution sui~ so as to may be ad~ptcd initial values :'grec with arbitrary of w and w.

202.]
202. th that of Let fondamental is to say,the th membrane. quation is

BOUNDARY. ORCULAR
us now examine more

267

of tlie character pa.rticukriy If ?t=0, iv is a function ofrouly, vibrations. to th centre with respect is synnnetrical motion circlus, Thc nodus, If any, are th concentric ~(~-)=0.G).

wliose

froin zro, w is a funcvalue dinurcut an integral of tlie nodal system tion of 0 as well as of 7', and th quation takes th form cos n (~-c<)=0.(2). J,(~?-) Whcn has Th sisting nodal of tlie into is thus divisible system circles represciited concentric J,(~.)=0.(3), and th second of tlie diameters two parts, by th iirst eon-

wherc are

is an

integer. The

Thse round tlie mdn

diametcrs centre of tlie

arc in circular

?!. in other

number,

and

uniformly ra.ngcd is arbitrary. position on. further vustiga.ted 203. Thc important

thcir respects nodes will bc in-

interal

formula

wliere

/t and

?' arc different

roots

of

~(~)=0.(2), equations by mcaus of th differential analytically may be verified and more but it is both simpler s:Ltisned J,.(K')-); by <7,.(~), whcre the thc more gnerai with to begin instructi-ve probicm, is not restricted to be circular. of th membrane boundary Thc variational equation of motion is

where

2G8 and thcrcforc

VIBRATIONS

0F

MEMBRANES.

[203.

In

thse

quations dispiacetnent is subjcct.cd. one ofits

w refera Let

to th actual with thc us now suppose

motion, that

and tlie

tbctical system cuting

consistent

conditions

to a hypoto which tlie

uormal

component

vibrations,

is exesystem so that w = M and

while

8w is proportional =p

to anottier from (3)

normal

function

v.

Siucc

c, we get

whcrc

/< bears

the sa.mc relation if thc penods, normal

to

that

/< Lcars rcprescnted

to u. by M and

Accordingly, hve diQcrcnt

vibmtions

In

obtaining

this conditions

rcsult,

we hve

boundary
actions

bcyoad quation

uga-inst

a-ccelcration,

in th

fundamental

no assumption as to the w!~t is impjicd in th absence of rif they existcd, which, would appear (3). is satisfied In ordcr identicallv, to evaluatc

madc

If in (8) we suppose and this we cannot intgral If equations iufer wc must thc

/c' =A-, th equation value of ~i~cfZy. a ratlier diffrent

follow

course.
certain contour

u and v be functions \7"M + = 0,

within a sa-tisfyiug + A: = 0, wc have

the

2G9 0FYNTEGRATED SQUARES. 203.] VALUES from 'U isderived that Let usnow thcorcm. suppose byGreen'a tliat so /C, uby varying slightiy ~tc. = = it v ~=~+0~; 8K, ~=~+-,0~ a/< K bic; substitutingm(10),wcHnd or,
if u vanish on the boundn.ry,

For

tlie application

to a circular

arca

of radius

r, we have

and tion

thus

from (10) on substitution to 6, with respect

of polar

co-ordinatca

and integra-

Accordingly,

if

and

/e and

?' be different,

an equation

first

proved

by Fourier

for the

case

when

Again

from

(12)

dashes

denoting

differentiation

with

respect

to Kr.

Now

97f) VIBRATIONS
and thus

0F

MEMBRANES.

[20:}.

0 as witli fixed s'hl~ bouu~laries, tlie to

cxpl'ossions
~=~

for 'L'fLIlcl
(~-)cos~+~7,

Inrom

to si 111 1)li fy tlic

wcfind

(~-)sm~j.

n)

Md

a suuihu-

equation

for

The

vn.).w. nf the work ~lone by t]m 't t 8~~ so t],at

impressec] forc:es cluring a ]iypothetic.vl if Z be thc ~=nnpresscd for~ reckoucd

clisplacement por unit of area,

lvhe"

0 aua

a.re

amalgamatod.

We

0 thcn have

~.rs~ constant pressure Z; thus

tliat the initial that assllmed

velocities

are n influence

,<,r. 1 1

20 4. J j No\v by thc difforential

SPECIAL quation,

PROBLEMS.

271 1

andthus

thc sommation

extchding

to all the

admissible

values

of/c, sti)t of the

As an example vibrations, of/b?'ce~ to spacc, variGS with respect constant timc. If Z= Titis cos of a small tnay bc takcn set mumbrane wc nnJ, ncarly to reprosent in vibration as bef'ore,

tnat wc may suppnsc a~ a harmonie function

th circumsta.ns roughiy of aerial wavcs. by a train

Th that, thcir

forced while relative

vibration none of
th

is of course
synunetrical

independcnt
normal

of
componcnts

It will
are

bc scen
missiug,

be :). if there grca.tly, especially in value ncar a.pproach bctwccn y a.nd onc of th sries of quauti. be vcry close, tlie cScct of dissipativo If thc approach tics forces must be included. importance mny vary 205. th nodal Th circles pitches depend of the various simple of tlie tones quation Mtd th radii or

on the roots (~)

= J,. (.) = 0. ta~eu of magnitude ues 10 a admissible n118SI e values then 1011 th in order be 0 of~p

If

thse

called ca ed z z",

(exclusive (~) z" w.

of 2;

zero)

272 are to bc found

VIBRATIONS by multiplying solution may thcn

cos ~) +

0F thc

MEMBRANES. by c a.

[205. Th.c

particular

quantitios be writteu

sin

?~)

cos

f~

6,'4

(1).

Th this there when

lowest

tone

of the ca does

group not

7t con'csponds vanish for any

to value

2!

a.nd since of r less than

in a,

case J,, (~) is no

interior

nodal

circle. <s)
a 2

If we put ")

s = 2, <7,, will vanish,

that which if we circles

is, when is the take tho radius root of th

r = a one

~n interior whosc !iodal radii are circle. witli Simi!arJy 1 nodal

wc

obtain

a vibration

(exclusive

of the boundary)

AU possibtc, and thus

the

roots of tho equation let Ka = X + bc a root by (14) 203,

= 0 J,, (~a) then ?'0 ==

arc

if For, t~ is also a root, re<

whose complex quantities, conjugate is necessarily so that theaboveequa-tion requircs product positive from tha~t either X or /t vanish. That X cannot vanish appears the considration cach term of that if rca were a pure ima.gmary, Now (~r), J~ (~) arc ascending sum of th /n=0, from or that th into th series series /tis the .7,, would of incapable Th that analytical only for bc positive, and We might functions thcrefore conclude be arrived of tho t!~o that at time vanishing. same result circular

real'.

considration

ca.n enter vibration. The equations

expression

for a normal roots

component zero). From

equation J" (z) = 0 bas no equal (7) and (8) 200 we get

(exccpt

205.]
whcnco would this uf we see
a.)sovanish

ROOTS

OF

BESSEL'S

FUNCTIONS.

273

tha.t

if for tti;Lt

<7,/ U~t

vanished ~atuc.

for thc Butin

same

-va.hie

of.s:, 200 va.lnc

wou)d

rcquu'e

H~ ttiu

functions

v)i'tueof(8) vtmi.s)i fm' ttic

iu (question'.

.3~ m~y bc found by into'pola.tion fron Hansen's t:).hl<j.s su f:t.r a.s thse uxtou) ) or furmuRc ma.y be catcutatcd froni t!tu duscunding sries of suceuH.stve by t)tc niettiod of routs th For t!)c it~portant a.pproximatiu[), cxprcHsiu~ dirccLiy. case of thc sytiunctricat vibrations (~ = 0), t)[C values J~ay bc of found frutn th fuliuwin~, ~iveti by Stokcn'~

20G.

Th

actu:)!

Yalucs

Th
con'csponding grcate;' indepcndcntly.

lutter

sries
tos==

is convergent
1. but 'J'!tC series thc

enough,
(1) firsb will root t!tb)e

even
sunice must (A)

for tho first

for ho is values

root,
of

t.h:m

unity; Thc

cn.Icutn.tcd t~cn from

:K'co]np~)ying

Stokcs'

pa.pcr,

with

a cither

sti~ht frum
mots

dittLit-ence

of

notation. or
i.s

It wHI
tlifTcreocG Tiu.s series is truc (10) M.

be secn
of

tho fo)'mul:H,
of )ngh erder

t)ic t:iUe,
n.pproxinmtcly

that

tho
7r.

successive

fur all vaincs

200.

of ?~ as

is vident

from

t))e

dcscending

Bourgct

the the

frcqucncies nodal circles.

in his tncmoir tab)c3 of gtven vcry claborate of th diiTcrent toncs and of tho rn.dii. of sirnptc
hn.s

Table

J3 i)t.cludes
9.

tlie

values

ofz,

whicii

SH.tis(y

J,.(.!),for~=0,l,5,s=].,2, BourRet, de !fo~ "M~mnircsurIotnnnvGmcntYibrntoiro onrwft~, thnt nn two tu., Hessefti La functioxn, 7'M); infinitc hns wlioso t. 1H(!(!. functtons donc ordurs ho. In ono of Tho

des j~nssnRo intf~r'~ thf'oron), t'y order

mf'mbrMtfs DonrHet cnn

eircu!Mrea,"

~inn.

1)M provod buticannotfmtithat in thti cnso

t)int ho implifs havo thf.' HMnn root, truc; frotn

of

JiOur

is pr<)t))t)))y howovcr, 1 or 2, it mny bo easity provud C)t!cu]&ti):t of ct~ss of

t)tofnrtnn])nf'f2()f). C'~Mt~. ~t<7. nitc intgrais and

Vt)I.

tx.

On th

num'jno~

dof!-

Rerics."

n.

18

274

VIBRATIONS

0F

MEMBRANES.

[206.

TAI!LE

A.

a 1
2

~fur.)=0. 7T '7'S

)df ~fM-(:)-0. (z) O. Di'.Difl: DUF. TTOI'.Ti .n.

'J!)~1 '!J!)!3 3 ~!I!1!l '!J!J~8 '!ln!i9 JJ''J ~~9

1'2~7
~0 :R:3 4'2111 1 6 .139 i;~3 l'()(lii:3 H)()~~ l l'(I OU3

r7~7L C) ~1: 3'75;H 4 ' fi~77 ~1~ (j'7fd!) ~S -7516

1 .) 7 S 10 t~ 12

8'2.I,j.1 3 1(1~`?.IG3 iin~~ 12-2.1G!)

l'll(I(): lUU.1 ,E Iww~ 1.0003

S~ J';rl3 irt'i~ n'7~1

Wth'n becomes nia.tcs thut to thc

is ratio

consido'ahic Forvcry
o<~)f~it.y, as

thc

calculatinn

of tlie

carlicr

mots

troubicsono. of

xmy

hig!) v:du(;s of bc scL'n frmn

approxit])u consittumtion

tend pitch of thc gm.vcst tonc ofn. very acutc sector tnust wit)i th:).t nf a. tong pamiiul to comcidu strip, whosc width ici c~ua.1 'idt)i of thc scctor. to tttC grcatcst

ABLR

B.

)t.=0

M==l

~~2

~=3 3 M=4

M=5 r

1 2 3 4 5 C 7 8 9

2'40t n'O 8-G54-1 11-792 14~:il 1 18-071 1 21-212 24-353 27-4U44

:~832 7'0!G fi 10-173 3 13-323 3 lG-t70 19-G1C 22-7f:0 25-903 29-047

f''I3.') 8'4!7 11-G20 H-7UG 17-OGO 21-117 2t-270 27-421 3U-571 1

G'370 9'7M 13-017 1G-2244 H)'.tl0 22-f!83 25-74!) 28-909 32-050

7~SG C) H-r'G.t l-t-373 3 17-G1G 20-827 21-018 27-200 30-371 1 33-512

8'780 I2'{M 1;~70() 18-983 22-220 0 2~-431 1 28-~28 31-813 34-983

20G.~

NODAL

FIGURES.

275

Thc 'vibnLtIon,

~gnrcs rcprcscnt ami the uumbci's

thc

more

afHxcd

normal modus Important r(jfut')'cd givc thu ft'cquenRy 182

of to

2 7 G

VIBRATIONS

OF

MEMBRANES.

[206.

the

gravest

cxpresscd cf six noda.1 diiuncters calculation Th th

tones

as unity, as fractions

with to~ether of thc radius t!)C frcqucncy

thc
of thc

radii

of thc circu~ar rcsult

nodes

mcmhra.nc.

Iti the case of a. rough

statcd

is the

by myscif.
eoi'rc.spolding

to tlio varions (!o not

fund:unental

modes scale, winch

of but may

to a, htu'momc bclong therc are one or two n.pproxima.iety Imrmouic relations bc worth notice. Thus circular monbr.mc x l-5!)t = 2-125 = 2-136 = 2-65:; ne:u')y, nearly, nearly; nodal diamctcrs

x 1-59.1. = 2-G57 2x Ho that give thc four 1-59~=3-188=3-156 modes

a consonant Th arca.
systcm

gra.vest chord.

with

oniy

would

of tho
m snch a.

membrane
a manncr

is (lividcd
that th

into
siga

nodal

serments th of modes

by the vibration

whencver changes which hve nndal the cal centre

vibrations

Ticdciscrosscd.

In thcre

those

of vibration

diameters of th

ofinertia

is Gvidcntlyno In th memitrane. of tbe centre

of displaceme])t case of symmfttriis propor-

t))c disp]aceinent

of inertia

tiona.!

to

an

which expression values of /c, sincc In all thc aymmct.)'ic:).l th centre of

incrtia

does

not

vanish

for

any

of the

admissible of

can)!ot vanish (.?) ami '~(~) is thcreforc modes thcrc of thc wc membra.nc. thc

simuitancousiy.

a. dispia.cctncu't

207. mcmbr.'me

Hithcrto to

ha.vc

but it nxcd solution of ther brane boundcd for a. membrane

fnnction.

hc complutc, that is vident

supposcd and th our

ctrcu!ar

a.rca

of

th

circumfcrcncc

virtually fur exampicsome cases of a memprohicms, circles. Thc by two conccntric cow/~e theory tbc second Besscl's in th form of a ring requircs thcory

to he on!y includcs th

Th

probtem

of thc

membrane

in

the

fonn any be

of mode

a scmi-circle of vibration to th

as ah'uady so]ved, inay ht.' re~ardfd uf whkh thc soni-circlc is capable

since !nn.st

app!ieab]c

207.] ]

FIXED

RADIUS.

377

In order to sce this, it ia on]y necessary circle a!so. complte to attribute to any point in ttje conpicmenta.ry semi-circle ttic to th~t whic)i inotiou o!)tn.ins at its opti.ca.l image in opposite the bounding diameter. Dus line will ttien requu'e no constraint to kccp it no(h).l. Simila.)' cotisidcrations to auy sector apply whoso angle is an atiquot pru't of two right angles. of th sector is arbitrary, the prohlem opening of Bess~l's fonctions of fractional order. may be soh'ed in terms = /3, th particular If the fixed radii are 0=0, solution is Whe]i th

who'c f7r used

is an intogcr. Wc Pcc th:).t if /9 bc an a.liqllt part of 7r, :un.t th suJutiou is inctuded /3 is integr:). amoug those a.lready for tlie complte cirelc. ca.su is when /3=27r, Intcre.stmg of uf whici) circle, cotti[)!(jtc tu bo nodal. which th corresponds radius ~=0 to th is cou-

Au pl'oblum str:uned

Wc

Lave
w = Pt7)(. (/f)') sui eoa (~ e).

When vibration new th modes

v la even, this without possib]e make their

coustraitit; lu appearance. V termintes, Thus, whcn

gives, the

as might

be

expected, but, -\vhen

modes v is latter solution

of odd, case is

for dcscc~~di~)g sries in nni.tc ternis. cxprcssibic

fact, in tlie so tliat tl)e ~=1,

The

values

of /< are given

siu /<<t =

by
0, or /M =7/t-n-.

278

VIBRATIONS OF MEMBRANES.
Thusthc circula)nf)th;sdiv!(tc ~nn thc ~xcd
hfn'mcnic radins scittu. into

[207.
equat

parts,

tunithe

MCt'ic'sot'tuntj'~

Int~e is~ta~y

e~suut'tho~r:LVust,tm)dc,thL!wt)()luoi'thn)u)nbrau' )nu)ncntdcHcctcdo)t thus.uuc sideof Ibis radins runuu'kubtcLh~t, ~=0 innkcs Lhu t,)tc
probtetn

its ut' t)t!Ui

:Lpp[i(.Lt,i"n
casier

c(~!i)ibriuta positon. L)m cuntit.t'aiuLtuthc buturu.

If wc t~kc

t~= 3, Hic solution

is

In this

case

thc nodal

l'asti

arc

2-n"='3' and tlie possible toncs arc givcn '=~T' by th cqun.ti.on

ta.n/ca=Kf!(4). To caleulatc thc roots of tan = x wc may assume

a;=(m-)-~)7r-y=Jr-~ whcrc y is a positive this, quuntity, we find

winch is smaU

\v!ien

is large.

Substitutiug wLcuce

cot

= JV

M,

1/1 J ~L x ~x(~~+r~)-'3--iiThis It will

Sy" -'3-15-31by successive

I7'/

is to bc solved cquatioti bu found tha.t rcadity = ,rl + Al J 2/=Y-'+~~+~~+~~Y3 ]ij

approximation.

X-5

10~

i-~

+.

207.]
so that thc roots

LOAD. EFFECT0F SMALL

of tMi <c = x a.re given by

279

whcro In n.nd a;:uitl

J\"==(M+~)77-. tho-c thc Hrst quildrn-nt in thc sccottd <p)adnn)t L:ui.Ba)-eopp"si~. is no root thcrc Th aftcr xcro since ))cc:m.sc Evcn ~ivc thc v:t)'tc th tana; signs is thus > a', of in case root

is noue

fii'stroot.iit'tc)-zro

thut)nrdqu:u1)-:L))t, thu sft-ic.s convoies wiMi a)i t)~ 2'-i.~0, 208. dunsity nsitierahic

c(.rrcspondin~to~=l. to suHicicntty

m tins of' Die

va.lucs of' ?~ it is fur I)i~hcr :t.ccur:Lcy, whitu 7r m'o l--t303, Th ~-tn:~ vahtcs of~ con)d )'e (h;sh-c'L C-4H-H-, &c. 3'-t.70U, 't'-i74.7, 5'4Mi~ Thc of tlic cH'cct on thc pcriods tTtC-nthnmu of n. sU~))t incqnd~y ))t~y hu invc.stig~tctt ux!LU)ptt;s to considctat a, point h:w(j tlie in hy thc th

circular

scvcrid xictinxt !)0, <'f ~hich gcnut'a.) heru IL wtH hu snH'iciunt Lcun '(.'n. to thc monhnuie s)n:dl io:Ld ~:Ltt:).c)tcd vector We sti)t
knictic

:Urca<)y case of a radius

wim.se

i.s )' wi)l t:d first the symmctncal from (~) + (~). (-M=0), types tlic prcseuec which of n~y Thc

hc supposai
cnergy

to :i.pp~y notwit.hs~ndiog 2'is (C) 20-t :dtcrcd t0 ~TT~ J~

p7T~ J;'

(~)

whcrc Th

P,

dnotes

thc

value

of~

whcn

thcre

is no !oad.

for dctcrmhmtc types are not ful)y unsymmctt-ic:U nnrt~a.1 Lut foi- thc prsent tttcy must purpd.se the unioadcd moubrane or n. ma-xinunn thc i-c.sultin~ so as to nmko pcnods bo tfdn load is th is to s:).y, so th.Lt thc cH'cct of the tliat minimum, a. !oad can ncvcr r~isc ~td lc:tst possible. Now, since greatest of tho !oad is tlie lu~st thc pitch, it is c)ca.r th:Lt th inthmnce diamctcr (it viz. xcro, whcii tlic type is such that a uod:d possible, t!)0 lo~d is nt which is mdiHcrcnt winch) passes t!n-ough the point to h:ivu two must bc supposud mcmbmnc Thc untoadcd ahtachcd. of thc is untdtcrud by tho addition couLCidoit pcriods, of which o;~

280
load. pcriud whun nud~l The other

VIBRATIONS

0F

MEMBRANES.

[208.

type is os ~rca.t as
r.t.dms

t)ic

diatnutcrij.

so that the fdtcration of chosen, which will ovidoiHy be t))c case possible, vecto!)-' bisucts thc futg!c bctwcon two <uij:LCcnt = 0, wc are tu take to T)tus, if correspond
'~=~~ (~)cosH~; i

is to be

so tha.t

(2) 204.

Of course, ncithcr circles, For

vanishcs, symmctricai

if r'

bc

such

t!mL

titc

lu~d tics on one of tbc tiodal

pcrio<t

is af!'uct.cd. contre ~ud pitc)) of

t)tc membrane.

cxampic, cxpt
vibrations

lut ~V Le at thn whcn )t=0;

whosc

(<))=!. isin!)uc[]ccd

It

is

J,. (0) th on)y load,

by a central

aud

furthclu

by (1) fil ~(~)~ ni ~(.)=-~(.), 0 ( ~no ) P

By(G)2()0 so that

the application of t)tc funrmJa a ktlowlo~c of rc()uircs oniy th va)ucs of' (2). whun For thu gravcst (.2) viuushes, 200. mndc thc value of J/(A-~) is -5190:}'. Whcn is cousidor~.0 abic,
~~o~)=2-7r~

approximatoly thc !oad in Tbc

systcni

so that
altcring

for thc

t!tc pitcb of a smaU

ca!cutatcd

tngacr components incrua.scs. irreguiarity in

thc

influence

of

it]f)uence
may be

fron

th

brcakin~ up form duc to tbe introduction hypcrbolic of suhsidiary symmctrical vibrations. In many ca.scs thc disturbance is favoured betweeu some ofthc by close agreerucut natural puriods. of 20!). a unifonu exact We will form. -341, -271, -23~, '20(!, -187, A'p. next how t])c natnral investi~ato vibrations are auected from by a s)ight departurc

obviuus

cn'ect

is

thc

the nodal disturbing formula Tbe most of 90. of nod:).! diamutcrs into curves

of th

metubrane

circular

ThoBUfoeedingTa!nosnrc~proximnte]y

209.]
Wttatcver cnuntion

NEARLYCI&CULAR

DOUNDARY.

281

ma.y Le th

nn.ture

of th

bounda.ry,

t~ sa.tisfies

the

whcre inny

/c is a, constant bc cxpundcd

to be dcto'mmcd. series

By Fouricr's

thcorcm

in tlie

whcrc sec that

~,w~

&e. arc functions must sa.tisfy

of r

only.

Substituting

i)i (1), wc

ofwhich

thc solution

is
M.~ ~.(~');

for, as in 200, Tbe gnral

the otlier expression

function

ofr

cannot thus

appear.

for M may

bc writton

~==J.J.(/<-r)+~(/<:r)(~,cos0+7?,sin0)
+ + J.. (/~) (.1, cos H0 + 7?, sin )~) +. (2).

For

all points In thc

on the boundary

M is to vauisb. mombranc r=ft+8?', condition th ~)' is radius bcing vector a small is

nearly function

case of a nearly circulai' Wc may constant. ta~e of Hence thc boundary

0=~[.7.,(~)+~(~)]+.
+ [' (~t) + t/ (/<:ft)] [J~ cos + sin tt0]

+. which Let is to hold good for aH values those therefore modes of of vibration wMcb

(3),

us considcr

first

are nearly

symmutrical,

for which

approximately

~=~.J.(~-). A)) tbc th type rcmaining of vibration coefficients can only arc differ to j~, since relativcly a little from w!iat it would small

282
bn, wcrc the

VIBRATIONS 0F MEMBRANES.
boundnry an exact circic. Hcncc if th

[209.
squares of

th!'s)n:)H(jU:u)titn'"b'~)mittr.L,~)!jf'co!ri< o L~. (~~) +

+.)-

'~o' (~)]
(~)

+ 'A (~~) [~i cos <9 + .B, sin <9]

[. co.s ?t<9 + 7~. si n /~j +.=(). (.t).

If

wc

intt'gratc

thi.s

cquattun

widi

respect

to

butwcc!)

titc

limits

0 aud

~7r, we ubtf).i)i

or
which s:unc This shcws as if

that th

thc piidt of tlie vibration is n.pp)-nx!)natu]y radius v~ctot- ha<! uui~rtnty its ~e~/t ~<e.
idiuws us

thc

to f'm-m :i rou~h csLimatc of t)tc pitch of whusc iij nuL cxtmva~mUy any mcnil.triutc boundary cjon'ratud. ]f o- dnote thu fu'c:t, su t)t:).t po- is t)tc )naHH of Ute whutc muniof t)ie gt~vost Lraue, tho frcqucjtcy {.une is approximatdy
t-csnit

2~ 2-40.i.x~(6). In mnltilly Thu.s ot~cr (-t) to the invcst.tc by eus y~, or sm altcred and ~P type thoi of vibration, int~-m-atc wc m~y as beforu

Witen

thc

vibration

bocomes tuorc question mcmbrMG truly circuler

thc npproxim~cly symmctriciil, Tlie nor))i;d tn0()cs io;- t). coinp)ic:).tcd. are to somc but tin' extcut Indetcnniuatc,

is

not

209.]
irrc~Iarity minatcn<'SH. no that values.

NEARLY CIRCULAR. BOUNDARY.

in th bonndru'y T!tC position will, ufthc in

283

Ute resulLin~ pct'io~s Lot us, howuvcr, snpposLw=~t~<7,,

tlie indcterrcmovc gnerai, nmst uudal difuncters ))utukcn, or minimum !n!Ly h:LV<! maxinium t!u).t thc approxiinato (/fr) cos ~(9),
th initia.! linc must

type

is

a.nd aftcrw~rds ordcr that

thc

this

invcsti~atc form may

how

bc ta~en

in

)io)<! good. bo!ng (4) ti'cated as small in compa-

A!l nson

rcmaming

coe~cients

with Jt.,

wc gct

froin

or

winch

shcws

tha.t

tlic

effective

ra.dius

of the mcmbra.ne

is

Thc
tcgraiin~

rn-tios
quation

of ~t,, a.ud 7?,, to A,. mn.y bo found as (10) a.ftar inulLiptic~tion by cos of~rcatcst
as to )ti:~

before

by sin )!0.

in-

But
is to bc

the
so

point
t:).kun

intercst
thc

is th pitch.
expression (11)

Tt~c initial
a maximum

line
or

minimum.
instc:).d

If we of we

refer liave

ta to

a, lino consider

fixed

byputtin~a t)~c dcpeudencc on a of

in spacc

the

quanti

ty

J
which may

r~eos~(~-a)~,

Jo 0 8r aiso bc writtcn

cos' v (B

u) clfl,

284
andisof th form

VIBRATIONS 0F MEMBRANES.
J. cos~ fx + 2Z?cos t/x sin va + (7sin'x, of nodal tl)e a. Thure are

[209.

2?,
admissible

(7 buing

indepundcnt for tho positions

thc

period ofon set set. Thcr

a. maximum, and bisuct the angles

diamctci's, othci' a minitnum. tlie

two according]y one ofwhich makes T!)c- dianietcrs of thc other

bctween

diameters

are, howcvcr,

cases

whcrc thc S/'

is

th nortna! expression
constant,

modes

remfun

InJc-

tC!'minatc;,witich happcnswhcn t'a. This is t!ie case whcn tional or in system pondicular this onict' ing thc to cos ot))erwor<]s corrcspunding dinmctcrs) For thc exa.mpic', hnund:u'y to )t=2 would

if wcro

wcrc

(12) is mdcpendent or whoi is pronorto cos 2~, proportional

ufapproxunation. to !t=l, must coincide witit one of ellipse, and tlic pcriod.s will be diircrcnt Wc hve SGGn that tho gravcst

uodal s]ight)ye!liptica),thc of a pair of pcr(that consisting ho arhitrary in position, at Icast to But th single diamctcr, con'cspondth axes principal for thc two a.xes. of

tone of a membrane, whose houndary is is ncarly the samc as circular, appruxhnately that ofa mcchanicaHy simil.'t.r membrane in the form of a. circle of tlic samc mcan radius or area. Ii' thc arca of a membrane hc thcre must bu some furm of boundary for wltich givcn, evidenHy th pitch tonc) is thc gravest and this (of thc principal possible, form can he no othcr than the circle. Ju th case of approximate an analytical demolistration thc circuhu'ity may Le givcn,ofwhich is an outhnc. foUowlng The gnerai value +. of~ +J,. being (xr) (~cos~+J9sin~) + 7? that (1), arcsmaM M vanishes

210.

~~=~1,<(/<) in which for the

to rc]ativcty wltcn ?' = ft + 8r,

J. (~)

prsent purpose tliecoenicients~ we nud from thc condition J,

J.'

(~a)

~J,

(~).

(~.)'

+S

[(J,(~)+ if ~cos~+/3~in

(~)

8?' + .l~eos

?~ + J3, sin ~)]

= 0.

(2).

Hence, ~'=

~+

+ ~cos/+

/3~sin/i~+

(3).

210.]

FORM

OF

MAXIMUM

PERIOD.

285

from

wLieh bc

we soc,
)nc:i.n

as

hcforc, <(/ca)=0,
radius

that

(}uantitk's

nc~1cct(~, :L ctoscr

th proxima-tioti ontur to obtailt

is

of tt)C small squares 01- that to tLis ordcr ofa.p:).]so thc L'Huctivu radius. In
\ve

if tlic

n,pp)'oxi)n;ttiun

<h'stdutcnniuc aud

and

~o

tcgr~ting

l'y multiplyin~ butween <hc limits

(2) by cus~ sin?;~ 0 aud 277-. Thus

~1~ inthoi

286
sothat

VIBRATIONS

0F

MEMBRANES.

[210.

T))C

<-)))cst.icn

is

i~ow

us

to thc fur A:<(,

sign

of thu

i-hL-h:UKl

muinbur.

If M= 1, :L'~t - Le wnttcu

vani.shcs

approximatdy

by

(7),

.since

= 0 n~u-)y. case ,(~) In the p~sc.nt iu b~ Lhc case, .sitjcc thc term fjuu.stion cvidunUy an i~turatiun wifLont of th cirdc disphtucmcnt = 2, (M) ~UO, Whcu t)~ buuudary.

gcncml Tt'ns f~<'

in

and ,~== 0, as shou)d

rcprcMcnts mcrcly

in

th

f~nu

uf

is pnsitn'c For this

Kiuma.nti's

of M grever for mtc-gra.l values wn n):)y nvail ourseh'cs purpnsu jf'r<~e D~rc;c/N~<~e/i, "or< c~u~Lujn has a r<)"t for

than

2, whoi of a. thojron to thc

.!=

2-401.

in givcn th~t cft'cct ?t. Thc

ubithur di~reuti~l

(t)<hcr ttt:ni xo'(t) l(.'ss than may bc put lato thc furm

whilu

luitiaUy

J, und

J,' Lv

wcll (~s increa.sing

as

.) a.nd tliat docs

~~c positive. not tlie cca.se range to

Accorddo so

in'~v- "-Lc~ins -log~ from before .:=?:,

whieh

it is cicar

within

= 0 to

210.]

ELLIPTICAL

DOUNDARY.

287

.3= M, ncither tha.n 2'4U~,

,7~ nor

can

vanish.

And

siuco

t/~and

J'

are

hothp'it~u:]tii~~M.itJ'('vsth~t,ii5anittt~gc'rgr('atcr

all va.ni.s)), 0, c)t.sc of nny mumbranc

f~t is positive. f~ is gycn-tcr

\Ve ccmdude than winch ci)'cu)ar

tha.t, nn!cns shews tlmt outHnc, tho

c' /9~ in th circle

of appmxim:)tc)y

ofc~na.tat'cacxccedsthocu'cIcoi'ctjualpiLtdt. Wc ttave seen that a good

may

cstimatc
bc

of th
oLtaincd

circulai' matly but by tucims cn't-'etL'd.

~hosc

monbranc of

c~nat.iott

Wc

will

sciai-axis

apply is Tt! anfl ucccntricity maj~r ofthubonndaryis

(~) a stil) ctoscr tu tilis method

of an npproxiarca a!onc, frutn its ntay Le approximation pitch case uf an ellipse, e.

thu

'J'!mp(;)ar

quation

In which

the

term

coutaining

e* shouM

bc correct. m tei'ms of c and the arca o-

Thu result Wc have

may also

bc expressed

and

thua

from tricity,

wLidi whcn

we sec how

smal!

is the

influence

of a moderato

ecccn-

the arca, is givcn.

288 211.
j)or<'h\!dar,

VIBRATIONS Whca thc

tbcpr~b~'i.'t

0F

MEMBRANES. of a. membrane is ncitbcr

\iL)'.i<i~ bc cvc'reotnc wltliont

[211. straight
prc.-cnts th

nxed

boundary

ofdck't'i-'i'jh.~ir.

difficultics intro()uction

witich

not could iM gnera! not hit!)crto of functions

discusscd

or tabniatcd.

of an uttiptic mnst bc ma()c in faveur bonndnry partiat exception thc i)npm't)).ncu of t-bu probton of t))i.s trc'ittisG but for thc purpoHcs thc i))tr(j(hK;t.ioti of compHcatcd to warrattt is scarc<)y sufHciunt invusd'J'h(jr(':K)uri8thurnf"ru!'cfurrc<ltot)K'()ri~iu:d an;dy.sis. that to n)C))t,U))i l'f't'o It will bH su~Hc'icnt ufM. ~!athi(;u'. l, gatx.n t))C ho]as. Solubtc
solution

txjdtd

systetu

is composcd

of

tl)0

confocal

cUipscs

a.nd

hypcr-

cases

m~y

bc

invonted

by

meaus

of

thc

gcncnd

!o=~l.J.(A:r)+.tFor exa.mp!c we might

?~ =

(.l..cos~+7?,.sin~).7,.(~-) take
(~r) X. J, (/<-r) cos

+.

dif1!rcnt attacinng thc to which bonnd:u'y U.scfui theoron thc tion diffcr iixcd of infortnation 88,
whicb

and

v:duc;s solution ~ay

aDows

to will
sonictinics us

X, trace thcu

thc

vfn-ious

forms

of

app~y. bc that obtaincd from th

to

ofa bouudary vibrating because tbe ncw state ofpitc!), from the o)d

provc niembnmc

of any contractioa an lvatnu.st cause to

constraint. linc and

becomc

Springs, th

mcrc)y by tbc wlthout incrtia,

of thin~s may bu conccivcd introduction of an additional arc

ils

s~ppnscd
cquitibrimn

to

urgc
position,

th

of

towards boundary prnposcd stin'cr. At to bicorne graduaHy until more rapid, tbcy approach

c:).cb a

stcp

thc

vibrations

linut, t)io

infinitc of

stiH'nuss

"f thc

nprings

and

abso1ut,c

to corrcHponding (ixity of thc-ir points

application. hve th H:unc For bctwecn Umits scribed In th instance,

Itisnotn(.'c~ssarythat as thc l'est, dcnsity the

or cvcn

p:n'tcntoffshou!d at a.)L at~y dcnsity is intcrmcdiate circles. for rcsult of the doser circuni-

tboseofthe won!d circle case arca bowcvcr that of thc to that

of a reg~dar pitch polygon and circmnscribcd inscribcd bc obtained arca substituting to ttic according tbc radius Is l'OaO, by

ofequa~

crp)al

th ratio of hcxagon, circle Inscribcd oft!te 'I.intni)].]M8.

of 210. of the circle the tnean

so that

2UJ

MEMBRANES

OFEQUAL

AREA.

289

by so much cent. Li t.he~iic w:ty we migh~conc)'dc(.h.').tthesect.<)rcfa circle of G0 is n graver form than tlic equilateral triangle th chord for th arc of th circle. by substituting Tho

of the

two

limita

cannot

differ

from th truth

as 2~ per obtained

table the relative in certain following giving frequency calculable cases for th gravest tone of membranes under similar mecbanical coadi fions and of equal cn-e~ (o-), shews tho effect of a or less departure from the circular greater form. CIrcIc. Square. _1 Quadrant Q f. 1 of~circle. 2-404.=4-261. ~2.-n-=4'443. 5'135./ f~?.~=4.~i~ ~s 45w1

Sector

of a circle

60.

6-379

A/~=4'616.

Rectangle Equilatcral Semicircic.

3x2. triangle.

/13 A/7r=4'G24. 27r. ~/tan 30" = 4~74.

3832A/~=4'803. /5 tna-ngle.J ,} 'T'~2 ~y~ 7!-A/~= ~r1/ same = =~'967. 4'067.

2x1.1 isosceles R~ctangle Right-angled 2 x,l. Rectangle 3 For former For 3 3x1. 1

Rectangle

-.) 5-736. G u, thc

if a square a.nd a. ch'c]c have thc instance, is th more acutc in th ratio 4-443 4'2C1. th circle th absolute frequency is

area,

In linear

th

case

of similar

forms

th frequency

is inverscly

as tho

dimension. of th frce vibrations of a membrane was thcory considered His in th succcssfut)y by Poisson'. l, thcory of th rectangle left little to be desired, but his treatmeut 1 Af~m. (le r~e(!(MMt'< t. R.
vm.

212. first case

The

1829. 19

~0

VIBRATIONS

OF

MEMBRANES.

[~312.

of

the

circular

membrane

vas
~~di.mortbc~'m~r,

restrictcd w~

to

th
-.ut

Rymmetncfd
-h i'c

vibrations.

Kirci-.h"~

dimcult, Ctebsch'a

problem y/~op-y memt)ranc

ofthc

circular

plate

(18G2) givcs q/Y~ thc efFects of stiil-ness and oi rotatory circular induding left be sccn that tticro was not much It will thercfore incrtia. aircady tlie mcmon- of Bourget ~evertheless to Le donc m 186G; accomof thc problem discus.sbn uscfui to contains refon-ed of which thc whoc Dumcricid results, complte by very paHicd wcrc not nc\v. howcver 213.
of various

publisbcd thc gnerai

in 1H;)0 aud thcory of tho

In his
m~terials,

cxpcnntcnta!
of winch

mvcsti~tions
papcr proved

M. Bourget
to hu as good

made as

uso any.

ofti'c snperHuons in wato-, and aftcrTonova! Tl)c papct- is immerscd of woud wbose ~noisturc upon a framc papcr is piac~d by blotting of th Thc contraction with coatcd bcGn havo gtuc. prcviau.sty edges faihu'cs but many tension, thc ncssfu-y in drying produccs papcr Evcn is cbtaincd. rcsult met wUlt bufurc a satisfactory be
mny

a wcll

ui cottsidci-abic pt-ecautions refiuircs of thc in pitch in consc.tuencc variations Uabic to grt use, bcing are cxeltcd 'i~hc vibrations of tho atmosphre, niuisturc varying to tiavc a scrics procecdiug of which it is necessary hy organ-pipcs, to thc cyc of pitch, :uid they arc mado vident by sma!! intcrvals If tho on H'c mombranc. scattcrcd sand of a littic means by on th thc s!uut accumultes be sufHcicntty vibration vigorous, wit)~ more or less prcciston. whosc fortn is thus dcHneJ nodal lincs, in thc tension shcws itsclf by th cire-les beeoning Any Ine'jUidity

strctchcd

mcmbmne

elhptic. Th
A

principal

results

of experiment
cannot vibratu

arc the foUowing

in unison with cvcry sonnd.

circult-

membrane

It tliat

can

with sounds itself in unison ouly place Is gcnt)y tapped. Iicard whcn th membrane Indicates, tho lines
souuds.

more

acute

than

As theory
aud Icss intervais,

thse highcr are

possible

sounds

are

separated

by Icss

thcybceomc. formed above distinctiy or Mow to rcsponse confusion cnsues, thc membrane altcred, in
supposai, a continuons

Th certain and
re.nains transition

nodal
deunite

oniy A littie

when

d~e piteli
un.aoved. from one

pipe is decidcdly Thero is not, as Savart System of nodal Uncst') ofthe

auother.

213.]
Tho
cu'c~os an

OBSERVATIONS

0F

M.

BOURGET.

291

nodal

Unes
~~Tn.tors,

arc
an

circlos
~ipory

or diamcters
rcrju~-f~,

or

combinations
tvh'ju

of
thc

ITo~'cvcr,

number

of diamcters

excecd.s iniddie

toward.s t!ic eonfuscdly are not well dcfincd. The Bourgct sidcr that sition capable might forccd borne wavcs gcncra.1 in thc case of the rcsn)ts vicws sa.me

to two, thc s~nd tends of the membrane, and

hea.? itself the nodos

laws
square

wcrc

vcriHcd

to th of be. and

conmembra.ncs'; of theory arc Elecisively established in oppoof Savart, who hc!d that a membrane was to no any sound, tierc remark that docs in not an secm to a membrane origin matter the in have what been motion its distinction pitch between

Ly MM. Bernard a.nd these authors

and

i'<jspondin~ But 1 must free vibrations Whcn tLcir

in mind. having

is set

suniciently by aerial

the vibration is orgau-pipc, not that th membrane asscrts, propcriy Theory spcaking /(j;'ce~. is only capable of vibrating with certain denned but frcqueneieH, that it is on!y capable of so vibrating When however th j~'e~y. of th force is not approximately to one of th period equal natural the rcsulting vibration periods, mny be insensible. the sound of th pipe was two or three cxpcnmcnts octaves higber than t)~e gravest tone of th membrane, and was ncvcr fnr from unison with eue of th sries of over aceordin~y MM. Bourget and Bernard tones. made th experiment under Savart's Whcn they sounded a. pipe lower in pitch than th gr~vest tone of th membrane, rema.ined nt rest, but was thrown into vhment vibration was approached. titc membrane,
cxperimcnt wa.s

In

more

favourable

conditions.

somew!~a,t tlie sand as unison

than th pipe was decidedly higher ~and returncd A modification of the again to rest. a pipe about a tliird higher madc by first tuning whon until above in its natural its tension that of tlie had condition. increased Th membrane to of sumciently the process

So soon

as the

than was bring

th then tbc

membrane heatcd pitch

th pitch of coincidence fe! cooling gradually manifcstcd itself motion of th sand, which at the by th violent was scnsihiy at rest. and end of th experiment bcghmiug M. Bourget found a good agreement of thc v:).tion with rcspuct to t)]C radii test wns not very prccisG,
~n;.
in consquence

During pipe. and the point

between

and obscrthcory circuler the nodcs, though tlie scusibic of width of

1860.

C~tt'w. M.

44947f,

~:)3

S93
the tones M.
bands

VIBRATIONS

OF

MEMBRANES.

[213.

of sand; but thc deviated considerahly of tlie memoir Frcnch suggcst

committee Bourgct's

of thc various simple pitch Th estimtes. from th theoretica.1 on to report appointed Acadcmy relative of th want explanation t))at It should also be remcmbered of perfect HcxibiHtya supposition as th closely a approached comparatively cause disturbing force of th on by small is tlie an ordinary force. But resistance of than

perfect fixity of th boundfu'y. the on th thcory procccds condition of tbings not at ail membrane sti-etchcd with

th most important perlaps thc air, which aets with much grcater on a string or bar in consquence The at mode gravest ail in points of vibration, thc same

a membi-a.ne

from tlie differcntiy highciof air from one side grca.t a transference

during direction, modes, which

cxposcd. large surface is which tlie dtsplacement bc affccted very might would other. not roquire so to tlie

CHAPTER

X.

VIBRATIONS

0F

PLATES.

to Green's method th quaaccording tions of eduilibrium and motion for a thin solid plate of uniform material aud constant we require th expression thickness, isotropic for th potential of bending. It is easy to sec that for each encr~y unit of area symmetrical if p~, Thus, for V will be cjiergy potential function of quadmtic bc tlie principal radii the is a positive thc two principal of curvaturc, homogeneous curvatures. the expression

214.

IN order

to form

of which J. must be positive, and constants, less than unity. Moreover if thc matcrial /n inust be numerically be of such a character tha.t it undergoes no lateral contraction when a bar is pulled must out, the constant vanish. This of information amount is almost ail that is recaured for our where A and purpose, statcnicut mcans nncd. From Thomson and Tait's -Mra~ content ourselves with a mere may thcrcfut'c of tlie relations of th constants in (1) with those by of ~hich t)io elastic of bodies are usually deproperties aud wc

arc

Philosoplty, G30, 642, if b be tlie thickness, that, 720, it appears modnius, y Young's thc ratio of latcral eoutraction to longitudinal and elongation for V is when a bar is puited out, th expression

294

0FPLATES. VIBRATIONS

F314.

If Mbo
of tho plane

the

small

dispiMcmcnt

at tlie point plate of tlie pta,tc arc ?, y,

wliusc

to thc plane pcrpendicular in tho coordinates rectangular

and

thus

for a unit

of area,

wc have

which plate. 215.

quantity

bm) to

bc

integrated

ovcr

the

surface

(~9) of th

We procccd noticed <o~ only on th th that

to find the tlie

variation tcrm

of F, but in V,

it

should

bc

prcviously reprsenta pendent

second

uame!yj< is thereforc

cMruct~tp'e of the p]:ttc, and state of thinga at thc edgc.

P,P~ de-

so that

ve

have

to consider

tlie two

variations

following comparison to thoso who wish iinvo trouble Youae's ~"(~) Ratio =~ (Clobseh Poiasou moda!uB=F

1 Tho

uscd of tho notations the oriH'H'H to conault

by tho principal mouuira. 'hx (TIiouison)

writers

may

(Clcbseh)=~

(Thon)aon)=:

~K+~t

(Tbomso!J)=? of latcral aud MiituaeJ contraction

(~rckLoS

tmdDoDkiu)=2~ elongation

(Hirchhof).

to longitudinal (Thomson)="~ to Lu and

Douldu)=<r this ratio

(Thomson)=~ Werthuim

(Eircidtoff).

215.]

l'OTENTIAL

ENERGY

0F

BENDING.

295

Now

by Grecn's

theorem

in which

f~s donotes with

un

clment

of th

boundary,

and

dnotes drawn

diH'crcntIation outwards. Thc have

respect

to thc normal

of tbe boundary

transformation

of the

second

part

is more

difficult.

Wo0

Tho tlie furm

quantity

under

the

sign

of integration

mn.y be put

into

'wherc and tlie

is tho

angle

bctween on thse, the

aud

tlie

normal side

drawn

intgration Using

boundary.

right-ha.ud wc (tnd

extends

outwards, the round

If

ve wc from

substituto tlie

ior ~8w

f/;<;

~8M; thcir <<y

values

in

terms tZ~

M~

quations

(sce

FI~. 40)

29G

VIBRATIONS

OF

PLATES.

[215.

wc obtain

Collecting

and

rearra.ugmg

our results,

we ~Ind

r~ -s-~

f~w.(~

+(1-~) "'(7~\

.cfw jcos~sm~ cos8smO \,t/y'

<~M\

'TZ)\ (cos' 8 +(eos~-sui~) 't~/y/J + J~/i. ( f~- ~'C7"M + (1 (' ~) cos' ~~M d~' .f~w -i- sin" a~

y 1

Tliere
motion.

force

in forming difficulty If p bc the volume aud Z~ deusity, c?6', acting on thc c!cment

will

now

bo no

the

of equa-ticns the transverse

215.]

CONDITIONS

FOR

FREE

EDf.E.

397

1 8F-ff~8w~+f~wSM~M=0.(7)'
must bc true wha.tevur which v:u'iation!U quation, gcno-al of ttic system) mn.y with tlie constitution fmiction (consistent ot' of thc Cidculus bo to be. Hcocu by tlie principles supposcd Variations is thc

at evcry If

point

of thc of

ph(.tc. thc pMe va)ucs bo froc, therc of 8w and is no restriction a'id on th th

thc cdgcs

hypothctic&l

bound~ry

thei-cfore

cne~cicnt.softhcsoquautities Th conditions tu Le: s~isncd

for SFmustvanisIt. in th expression at a, frcu cdgc arc tlms

If th whole and If
sccoud

circumfercucc

of the

pla,te be clampcd, conditions hut~ia cl~a

tins case bc

5w = 0,

cln

= 0,

tlie satisfactiou th cdge 0

of th

of thc boundary ~=0,

must m

is already tn-bttrary.
s~tisfied by

sccured. Th
w.

bc 'supportcd",
cqua.tious (9)

by getting may be simplified bound:n-y quations coin tho use of Cartesian involved lment rid of th extrinsic of tlie to th normal the axis of a: pM~Iel ordinates. 'l'aking curve, wc sec that we may writo buuuding 216. The

Aiso
CoinpMe 1G2.

Tho rotatory

inortia

ia l'cre

uc~locted.

[216. whcreo-is a,fixcdaxis coincidiugwith thc tangent at t!t0 point (I"w T b underconsideratiou. d l '1 Ingenci'a.l-diH'ei'=ih'om C12w.1' M~* <M' Toobtain th relationbctween them, we may proceedthus. Expn.ndw by Ma.cta-unn's thcorcni in ascoudingpowcrsof thc smaUqua.ntities n and o',and substitutofor Mand o-thcn.'valueslu terms of a, th arc ofthc curve.
VIBRATIONS 0F PLATES.

298

Thus in gnera). fF~ ~"w ~<; Ajj ?~+ ?:o- + -r 0-'+ w= + + o- + (/)!~ ~0-~ ~oft whUc on th curvc o- = s + cubes, == s'" whcrc /? Is th + for pointson th curve, radiusofcurvature. Accordingly

and thcrcforc

whencefrom(l) 2 Io ~"W. 1~!0 0~~ ~"tc=-+-+. v (3). (~' p We concludcthn.tth second bouoda.ry conditionin (9) 215 be put iuto tlic form mn.y

In

the

sa.mo way

by putting

== 0, we sec th!).t

is quivalent of M nd

to

cht cl~

wherc Th (h'st

it

is to be undcrstood boundary conditiou

that now

the axes becomes

cr a.rc Rxcd.

If

wc

apply

thse

Ct~uations

to

th

rectangle

whose

sides

arc

216.1 pa.mllol sutisncd to the

CONJUGATE coordinato cdges axes, pa.ra).iel

PKOPERTY. wc obtahi to as th conditions to

200 299 bo

:).long tho

In

this ca.se the

distinction

betwecn

radius other rosults out

of curvaturc, pair of edges may

is inlinitely are found by be obtained cquaUy well transformation.

and p, thc s disa.ppcars, for tlie conditions Th gt'ea.t. Thse x aud y. mterchanging o- and from (0) 215 directly, with-

th prelilninary

Auy boumi:ny

two

values

conditions,

of w, K and arc co~x~e,

con'csponding that is to say

to

thc

same

this to prove In order bo diffrent. tha.t tlie periods provided ha.vc to diU'erentia!. from thc oi-Jiuat-y (3), we should quation Tins ia the method retrace thc stops by which (3) was obtaincd. for th ch-cular dise, but it is much aimpicr !~dopted by Kirclihoff Mid more direct to use th va.rin.tiond quation

in whick displacemcnt
symmctrical

w refurs
consistent funetiou

to

the

actual with w and

motion, th nature ~M, as may

and

8~

of tlie

%ti arbitrary SF'Isasystem. to from 215, or

of

be seen

from

thc general

character

of V ( 04'.)

300 If we hn.vc now

VIBRATIONS suppose in tlie

0F first

PLATES. place th~t w = :t, 8~ =

[217'. wc

~~=~~f:tu~;
and i)i Hkc nia.tuier if we put w = v, 8~ = u, which wc are equally oitittcdtodu, gr=~f~~s', '\vbencc

Tins boundary,
'\vltolcori!ipa.rt.

dmonstration and whcthcr

is valid th cdge

wl~tcvcr

may

be thc

form

of th in

be cla.mped,

supportcd,

or frec,

As

for

thc

case

of mcmbmnes that

in thc

the

to prove (7) may bu onpiuycd from but tins is vident ruai; 218. For thc application

la-st Chapter, admissible v:duus

equation of arc

physieal

cousidcrations.

to dise, it is necessary of polar coordinates. the quations Taking by means express to titc ccnti-c uf tlie dise as polo, wc hve for the gcncral uquatiun arca at ail points ofthe bc satisnud to a circular

To

cxprcsa

the

boundary

condition

( 21G)

for

a frcc

~d~o

()-=(t),we!m.vo

= radius

of

curvatnre

M;

and

thus

AfLcr to.

tl)C diiTcrcut~tions

are

pc-rfurmed,

r is to Le made

cqu~I

218.]
If w bc cxpa.nde<l

POLAR

CC-ORDINATES.

301

in Fourier's
w = ?t~ + +

series
+ +.

each

term

sepM'atcly

must

satisfy

(2), and

thns,

since

!~<x cos(M0a), /r~ d1' ~~p+r~'r~ (l'zo"

~=0

l~A 1 +

,/2-~r~

3-~ US
0

Il\Q' (3).

(_luu"(2 nt

The

superficial

diferentia.1

quation

may

bc written

(V'+~)(~)~=0,
which becomes for the general tei-m of th Fourier n_' -a )-' will expansion ,Y, = K M by ndding solutions of M' .1 ~_z' ?" )'f~' f +--T-i+~('n+'' ~f~ shewing togcther, that with tlie value complete constants arbitrary .1~ 7 ?'~)' of

bc obtained th genera.1

preHxcd,

with the npper sign is the samc as that which equation and of circular membranes, in th case of th vibrations obtains that thc solution wc conclude applicable as in the last Chapter of r is ce J.. (/o-), the second function in hand to thc problem The bcing hre inadmissible. same
x ,7,,

In tlie
nigu is Wn

way the
(~r), whcre

solution
t == s/

of tl)c
1 as

quation
usual.

with

the

lower

The
M), =

simple
cos ?t0

vibration
{o( J,. (~-) +

is thus

/3~, (~?-)}
quations
which

sin

{'yJ,

(~-)

SJ,.

(t/<?-)}.

Th values a of

two

boundary and the

will
must

dtermine bo given quations 8,

tl)e to

admissible the ratios

values

8. ~3 and y that we must have and thus ' may

From

the a

form

of thcsc

it is evident

/3 = 'y in the

bc expressed a) (J, (~-) +

form cos (~ e).(5).

t., =

P cos (~

(~))

302

VIBRATIONS

0F

PLATES.

[218.

As ni tlie

th

case

of a, membrane

the

nodal

diametcrs

othenviso

symmetric~Iy dcnoted by arbitrary,

(UstribuLed

system round

is composed tlic centre,

of but

cos(~-a)=0 togethcr with tlie conccutric circles,

.(~, -\vhose ctiu~tion is

J~-)+XJ,(~-)=0.(7). 219. In order to dctcrmiMC Whcn tue ?L a.nd edge is we must we ultrcduce obtain th from

conditions. bouudary (3) 218.

free,

in which by

use has

been

(/c-), J,.(~)'). numcrator from tLc by writing minator ma,y be dcrh'cd wliose is obtuiucd ofXthc of cquntiou BycHnunatioa of /c. vidues tlic admissible Whon = 0, tlie rcsult assumes Jn('A:a) 2(l-~)+~~+~y~=0.(2). This, of course, could M from ttic bcgiumng. The calculation of ha.ve beGH moro easily obta.incd aj simple ~(~'t) form, rt viz.

tl<e diU'crcntiid quations on th right In e~ch of th fractions nmde of

satisncd Hie dct)oin place roots givc

/'9') by neglecting is trouble-

the lowest

root

'be cceted of uppropri~c some, and in the absence sries fur thefunctions .y,.(!). ~(~'), by menns of thc asccnding roots recunrsc ~n~y Le h~l to th semilu the case of tho higher Kirc))hoff fur thc s~ne functions. descend ing sries convergent finds ~L-+ tan (~Tr) 8~(8~)" = T
~~8~~(8~t)~

for each value of must tables

(8m)"

whcrc ~=~=(1-~)~=ry(l-4~) C = 'V (1 7) = 4~') ((1 -8, (9 4~) 4~) (f) + 4.8 (1 + 4~), 4~) (13 4n')] + 8 (9 + 136~ + 30~).

219.]

KIRCMHOFF'S

THEORY.

303

where~isanintcger. It
the

appears
of

by

numcrical nodcs,

comparison
n.nd (4.)

that
cxprc.ssGH

/t Is idcntical
a, law discovei-cd

with

uuinbcr

circulai'

that tlie ft-e<tucncics by CIdadni, of nodal diameters given number lowest,

cven or

approximately
uncvcn uumbers,

proportional
according

to figurcs with a eorrespouding the exception of the arc, with of conscutive to tlie squares
as thc number

of the diamGtcrs of (4), we when any be number

is itself sec also

cvcn that

or odd. th pitch

'Within is

th

limits

approximately froni A, provided is subtracted number addcd to ?:. This law, of which traces appear ill the following table, nodal th pitch raising by saying that towards may be expressed It is probable, of nodal diameters. circles have twice t!ic eScct have no two normal components that, spealtiug, however, strictly exactiy th same pitch.
?=0 Ct~ 1 2 Gis g:s'+ P. HiH+ h'W. A-tb'-(Cn. b o"+ ~=-1 1 P. hf"+ ~=3 3 w. Cil. P. w. W. cfts"+

of application unaltercd, that twice

/t

/t i 'cnr*'p.~

?t=3

0 1

C g'

C
gis'+

C
a/-

d
d".dm"

(tisdi8"+

d)sc"-

Th ofthe being Ch, and values higher,

table, extracted more important to bc assumed refer

from

Kirchhoff's

overtones The C' to theory

deMotes tlic

mcmoir, gives th pitch of a free circular plate, th gravest tlie heads under columns three
rcsults as Poissou's observed and by

as of

respectivcly frou calculated

A a 7)~)t!<s sign sign

Odadui

with that

Werthenu's

tlie

actual lowcr,

aud

that

it is a little

is a little pitch than that written.

1 Gis corresponds to (~ "t the EnK)ish notation,

and ~t to b natural.

304
Th l)ut th disercpancics not perhaps plate. 220. Titc radil

VIBRATIONS

0F

PLATES.

[219.
are considrable, to jrrcgularity in

between greater

theory and observation than mny bc attributcd

of the

noda!

cit'dus

in

and with (t; = 0) were calcuiated by Poisson, numbers The following results obtait~ed by Savart. expci'ime!)tf)J]y from a papcr by Strehikc', arc taken who made som careful meadise is taken as unity. suremcnts. The radius oftho
Obsorvnt.ion. One circle Two Iwo 0-67815 fO-39133 CnIeuJation. 0-68062. 0-39151.

symmctric:d by him comparcd

tlie

case

cu-cles.
circles 0-50107

~.g~~ fO-25631 0'25679.

0-59147.

Thrcc

~0-893GO 0-89381.
Thc would The
()! not

ca.!culated vary very

rcsutts

appcn.t-

to refur

to Poisson's were
of

value

of

but

little

if Wert.Itcim's
a,

v~luc
comparisou

substituted.
Kircilhoffs

titb~ givcs foUowing zro) with measuremcnts 7?~~M

by Strebikc

m~dc

theory on less accurate

dises. q/' ~'CM~ft)' 2Vo<~<M.

]cu!a<on.

Obser~tbu.

~==nP.). ?t=l, ~=2, ~=3, M-i, A=l 1 /t=l ~=1 1 /t~ 0-781 0-70 0-838 0-488 Q.g~ 0'783 0-81 0-843 0-493 Q.g~ 0-781 0-S3 0-783 0-7M136 0-82194 0-8.1523 0-40774 4 0-87057

~=~(W.). 0-78088 0-82274 0-84G8I 0-49715 0-87015

221. or of thc not, the

WLcn theory nod:U

is truly plate symtnctrictd, whctherunifonn and exporiment indicntes, veri~GS, th:).t tlie position diameters is fu'bitnu'y, or ra.ther dcpcndcnt only on which tlie

thu

is supportcd. thc pl~tc By varying (liamctcr It is any dcsircd mny be made nodal. place of support, is a.ny sensible othcrwise wlien t!)crc from dcpartui'c goncraUy of vibration, Ttic two modes whicli exact symmctry. originany, manner in
1 Pc~e/iH".xcv. p. 577. 185S.

221.]
in any tlie consequence

BEATS

DUE

TO

IRREGULARITES.

305

of

the

equa.lity

proportion and separated

without anected

ceasing with different

of pcriods could be combined to be simple are harmonie,

in now

of the nodal position limited to two alternatives. by rotation diameters uniformity

At the same time periods. diameters becomes or rainer determinate, The one set is derived from the other

composed of the deviation

haf the angle included between two adjacent tlirough of th s:nnc set. This supposes that th deviation from tlie nodal system is small otlierwise will no longer be and diameters of approximate circles at al!. Thc cause

either in thc material or in may be an irrcgularity the thickness or In the form of thc boundary. Th effect of a small as in the parallel load at any point may be investigated problem If thc place at which th load is attached of th membrane 208. does minate. not lie Th on types are made deterdiamtral to one of the types system corresponding the place in question, and for this type the period Th period of th other type is Iiiereased. circle, motion gnerai th superposition, normal of the of th with uniform arbitrary circular plate is and Th to a nodal tho normal

passes through is unaltered. The expressed most

phases, dtermination

by of the

amplitudes investigated. to correspond

components

initial arbitrary as in the corresponding problem the characteristic of th property Th is eit,her two other cases of

amplitude and displacements

already and phase velocities

precisely for th membrane by th aid of normal functions proved in 217. plate be easier in which the edge

is effected

a circular

or ~)o?'~ would than th preceding but are of less practical on in their interest treatment, of th difficulty of expcrimcntally the conditions account realising th nodal The gnral resuit that is composed assumed. system is of concentric distributed, circles, and diamctcrs symmetrically to all thc tin'ee cases. applica.ble clamped theoretical 222. by sand Wc have seen that in general with th dviations Chiadni's circles ligures as traced and diameters of usually be re-

theory attributed membered

agre very closcly but in certain cases

to irregularities in tlie plate. excited that the vibrations

are occur, which It must however

and free, speaking certain modification. two or J!. mnre normal

by a bow are not strictly that their are therefore liable to a periods It may be that under the action of the bow vibrations whole coexist. The component 20

306

VIBRATIONS

0F

PLATES.

[223.

motion althougli

may the

explanation in certain obtained Another manner

in virtue of tho external harmonie force, simple diffrent. Such an would be a little natural periods of th figures charactcr is suggcsted by th rogular be cases. of tho deviation may perhaps Le found in th cause

of are supported. Tho rcquirements plates WheM to meet in actual are often difficult cxperimcnt. theory with an imperfect this is so, we may have to be content comparithat a discrepancy but we must remember may bc thc f:~u!t son as well as of th theory. of the experiment in which with which we to solve th problem attempt who succccded is duc to Sophie have just been occupied Germain, led to differential but was in obtainiag tho correct equation, For a frec plate tlie latter erroneous conditions. part of boundary In Poisson's is indeed of considrable thc problem dimculty. des corps diastiques' et le mouvement mcmoir Sur l'quilibre to be as necessary mathematician that eminent gave ~7'ee equations 223. The first tht~t bas proved at aU points of a free edge, but Kirchbon* to satisfy thcm aU. It happons, it would be impossible in gnral an exception occurs in the case of tlie symmctrical that however, is true of a circular vibrations plate, whcn one of tlie quations of tho Poissou's to this theory pcculiarity, Owing identically. is correct, the error in his vibrations notwithstanding symmetrical was In 1850 th subjcct conditions. view as to the boundary who first gave thc two quations resumed appropriate by Kirelihon' the theory of thc vibrations of a cirto a free edge, and completed satisfied cular dise. 22~. bcen The correctness of Kirchho6''s boundary quations whero explaining bas

lie who, without disputed hy Mathieu", a dinEcrent set ui error to lie, bas substituted considers Kirchhoff's so He provcs that if M and u' be two normal functions, quations. that w=~cos~, w='eos~'< arc possible vibrations, thcn

m~. Crelle, tichenScttcibc.

de !4Md.

d, Se.

<t Par. dus

1829. CIcitihgowicht und die Bcwcgung cincr c]~-

t. XL. p. 51,

Ucber

~Z,~)f)-~t'.t.xtY.J8G9.

224.]
This the follows,

HISTORY

0F

PROBLEM.

307

if it

bc

admitted

that

u,

satisfy

respectively

e<~uations c* ~7~ =~, Since th left-hand member c" ~7~<t' ==~/V. is zero, the same must be true of

to Mathieu, cannot and this, according member; right-hand Luth u a.nd u' of th boundary bc thc case, uuless at ail points pairs of equa,tions satisfy onc of t!ic four following the

for a free edge, but pair would seem the most likely Since th first and third it is found to lead to an impossibility. Mathieu coneludes that the fourth inadmissible, pairs arc obviouely th condition must be those which really express pair of equations In his belief in this result hc is not shaken of a frec edge. by the for th free end'of conditions a bar fact that th corresponding Thc second would be

the

first

of which

is contradicted tuning fork.

by th

roughest

observation

of

tlie vibration The fact

of a. large is that

of quations any of the four pairs of the boundary in (1), it th would secure integral eau be made to vanish that the integral does not follow conversely is negatived in no other way; and such a conclusion by KirchhofPs other are besides innumera/bla cases in There investigation. would a.11 that is really in question vanish, thc integral which sbould be either at the bounda.ry appliaBCCs being that necessary although evancscence l'est, or devoid 225. .suMWteJ, functions tbf same Thc mny of inertia. vibrations bc of a whose edge is normal of

rectangular

plate,

the theoretically, casily investigated to a membrane with those applicable being identical If we assume is fixed. shapc, whose boundary

2D2

308

VIBRATIONS

OF

PLATES.

[225.

we sec

tbat

at ail points ~=0,

of thc

boundary, ~-n ~-0, ( 215)

~=0, of

whicit The

secure value

thc

fu~hnent

of p, found

conditions tlie. ncccssa-ry in c*o=~'M, by substitution

sbewing squence It

that

the

anatogy

to

th

membrane

docs

not

cxtcnd

to th

of toncs.

of the prnnary bero the discussion to repcat is not necessary to IX. It is enough in Cbaptcr nodal systems and derivcd given modes that if two of tlie fondamental (1) hve th same observe membrane, in tlie case of th thcy must ah-so hve thc same period are nodal systems The dcrived tlie plate. period in tlie case of i)i tlie two cases. idontica! accordingly The with freucratity of tlie value of
phases

obtaincd
ail possible

by

compounding

and arbitrary amplitudes tions of tlie form (1) i-cquircs

particular

solu-

no fres!t

discussion.

it would bave madc, of a ~M~w?' the nodes plate to say that seemed unnccessary which Cbladni's the ordinary ng'n-es, to do with bave nothing to a plate whose cdges arc frec. belong fur a snpportcd conditions edge is of the The realization Unless th contrary assertion I~ad bcen are required capable in practice. attainabic Appliances scarcely at l'est, and of sucb a nature of tlie plate t!)e boundary of holding Wc may axes. about that they give rise to no couples tangential thc against to be hc!d in its place the plate by friction conceive round it. circumscribed watts of a cylinder closcly are whosc cdges of a rectangutar plate, problem and bas for tbe most part rcsisted dimeulty, frec, is onc of grt is independent tlie displaccment that If we suppose attack. with that identieal bocomes differcntial quation of?/, thc gnral If we take t)te in Chapter Vin. with \vbich we werc concerned ends are frec. to the case of a bar whose solution corresponding and tbci'cfore satisfying 226. The 0 < o <

22G.]
when .c=U and when

UECTANGULAR

PLATE.

30!)

a;=~,

we obtain M well

a value as th

of?o pair

which

sa~tisfies

t!fe getierai
cqua.tiou.s

(liierenti~l

cqua.tion,

of boundat'y

:u'c a.pplica,b}c to tho cdges to y; but parallel for thc ottior p:ur of edges, namely condition boundary ~M f~t?

which

tlie

secotu)

CI;C ~+~~=0.(2),

(?) This shews that, exccpt in the

will case

be violated,

uniess

~.=0.

it is not reserved, after tlie manner vibrato tion, that may

for a frce rectangular possible plate to of a bar; uuless indeed as a.n approximawhen the Jcngth is so grcat paraHcl to one pair of edges to bo satisfied thc conditions n-t thc second pair of edges out
the

be left

of account.
cottstaut fk (which expresses

Although contraction is positive

thc a, bar case

ratio

of lateral out) sub-

to for

longitudinal every known

extension

wbeti

is drf).\va of a. few

for cxampicit Therc stancescork, very smaIL absnrd in the Iden of a substance is, so far as we know, nothing vanishcs. Thc investigation for which of the undcr probtem this condition is tilercforc not devoid of interest, the results though will not be strictly to ordinary applicable glass or meta.1 ptatcs, for which If tiie value of is about the normal 1 functions for a frce bar tiodcs, invcsth &c. dnote

in tlie substance, is comparatively

in Chapter vin., tigatcd vibrations of a rcctangular

to 2, 3, corresponding will be expressed plate by

n. pinte of mfttorial, for wluch is not xero, vibrato m tho of a bar, it would mnuner bo noecfiHfu'y to apply to tLe edgea conHtnutling couples to the p)anp of bondinn to provent tlio aasumption of a contmry pnraU! earvfttuTe. Tho oficct of thcso wouH bo to rnise tho pitch, und thorofora tho caloncouples intion founded on th type propnr to ~=0 would in fiomowhat give )t rosutt higbcr pitch tlxm the truth.

1 In M'dor to rnuko

310

VIBRATIONS

OF

PLATES.

[236.

In each of straight recta.ng]c. vibrations

of these lines Whc)i

primitive

modes

th

to one or parallel b = o~ th rectangle /a;\ ~=~u' a 1 "=~

is composer system of tho cdges of th other a squa-rc; aud the becomes

nodal

Il a

in any protlie same period, may be combined nocessarily harmonie. while th whole motion still remains simple portion, nodal curve will of Whatever th proportion may be, the rcsulting having necessity pass through th points detcrmined by

Now The nodal

Ict

us system

consider of th

more

particularly mode,

tlie

case

of

= 1.

of a pair of straight nearest cdge is '2242 by the th corresponding curve that

consists w = M, a [ ), from the lines parallel to y, whose distance a. Th points in which thse lines arc met primitive pair for w = u1 a,rc thosc must ('), vibration through iu a.ll cases on th bc taken equal and wlticli pass. equal, opposite

nodal

of th they If tlie

It is vident of tho but in square.

compound are symmetrically two (or, primitive

disposed vibrations with

diagonals

opposite phases we have amplitudes),

algebraically,

from th

which

it is evident which th and

that

w vanishes tlic

whcn

diagonal vanish along th functions,

passes through other diagonal we conclude that

origin. from follows tho nodal

a:==~, tha.t is along That w will also of symmetry of (3) comsystem thc

FiR. 41.

prises both the diagonals (Fig, vibration of a square plate.

41).

This

is a well-knowu

mode

of

326.]
A scccnd their phases notable tlie

CASE

0F

SQUARE

PLATE.

311

cage is when

the

amplitudes

arc

cqual

and

sa.rne, so that

Tho thc curves, in similar straight n first metical

cases,

most cases.

convenient

method

of

constructing

for which

is that M=const., employed of eurves Tho two systems (in by of a~j th = const., constants a )~J

graphically by Maxwell this instance a.rc

Unes) laid down,

represented th values

= const., an

with th sa.me common progression is obtained In this way a network cross an The execution diagonaUy. inversion of thc table given march of t!ie function of

forming diffrence which tlie th

arithin the two required

eurves re<[uires expressing follows

th

plan proposed in Chapter yllL, 178, result is as M~ of which th

M~

a::

M,

~:0

+1-00
-75 -50 -25 -00

'5000
-3680 -3106 '2647 -2342

'25
-50 -75 1-00 1-25 -1-50

-1871
-1518 -1179 -0846 -0517 -0190

by the above values of x (comside of th central line) and pleted symmetrically in the figure (42). tlie corresponding for y arc laid down system At the are deduced. From titcse thb curves of cqual displaeement and equal to 2 on th centre of tlie square we h:Lve w a maximum Th system of lines representcd on th further side outwards is th locus of The first curve proceeding a-dopted. th Th ncxt is tlie nodal line, scparating points at which w= 1. in curves taken Th remaining regions of opposite disphcement. The numerically 3. order give thc displacements 2, great1, of the square, occurs at tlie corners est ngative dispt~cement to 2 x l'G-to = 3'290.' where it amounts On tbo nodat linos of squnro plate, Phil. Angust, 1873.

312

VIBRATIONS

OF

PLATES.

[226.

The

nodal

observations fully worked

agrees pretty closely with t!]c careof Strehike Hia results, winch refcr to three in tlie following of glass, are embodied polar plates curve thus conatructed

quations: -40143 r= -40143 -4019 -017H + '0172 -OJC8J cos4< -00127) + -00127~ cos 8~, '0013 1

the centre radius

of the

square

parallel the c:deulatcd -41981, -4.200, whilo vector mefmm'cd a diagona.1 along e~culation -3900.

vector

being to thc

pole. sidcs

From of the rcsult

these

we obtain

for tire '4-1980, radius and by

(<=0) square is -4154!. Thc -3855, -38C4,

is '3S;')C,

rf'g~. -hfM. Yol. CXLVt. p. :<lf.

226.]
By crossing locus of points th for which

NODAL

FIGURES.

313

network

in th

other

direction

wc obtain

the

for that winch are tlie curves of constant is constant, displaccmeut Thc ~<c/t of thc vibratiou M-e nodal. ttmdc in wti!c)i the (Uagonals lu both cases. thc samc is (accordi ng to theory)

The

primitive

modes in likc

represented manner.

by w = ~ FIg. 43 shews

/.K\ ~t the

or :t) = M,

may be combined for thc vibration

)~) noda.1 curve

.(~(~<.). Th ii' th form sign of the curve is th rciativcly bc altcrcd. same to tlie othcr diagODa),

of the amhtgmty

314
227. application t!ie form passes whethcr x thc tion or to rcsultaut nodal tlmt of no into thc Thc ou any may bc,
that

VIBRATIONS

OF

PLATES.

[227.
for its dpend function. Whatever r = 0 winch wben the arc bc same period, to par~nd not

method

of superposition for)n of particular th modo discussed, straight of

docs
norma.1

jnst

vibra.t.Ion, must have nodal

approximately If the two lias

system

lines

still of

vibrations synchrouous Hjc sa.iuc ])criod, and the

bu

superposed, course gnerai

thc considra-

of its

may

tra.ced

primitive thc line normal vanish, brotllers capable Charles it to thc taincd a plate, assumed. vibration

point vibrations

th

hve

plate the sa.mc

of by mcans ca.)i bc nodal

at

w)nch

tho

sign.

To

dotcrmiuc

exact)y

a complete of th primitive compensation, knowledge and not mercly of th points at whicti functions, thcy in gencra.1 would be necessary. and th ])octor Young to have had thc idea of superposition Weber as appear of vibration, but it is to Sir of giving risc toncwvarleties that we owe tlie first systematic Wheatstone' of application, of Chiadni'a arc however figures. Thc results actually obto only very roughiy applicable form of normal function implicitly only compound an

cxplanation by Wheatstonc

in consquence of thc In place of Fig. 42 (itself, bc it remcmbcrcd, WIieatstone nnds for the node of th approximation) th inscribcd square shcwn Fig. 44. in Fig. 44.

This

form

of rigidity, of th pomt cular to

is rcally apptica.b!c, but to a. sti'ctched ch'cumfcrcncc p!:uie of th

to a, plate vibrating~ In. virtue so supported th~t cvery mctnbranc, is free to move n.lon~ li.ncs perpendimembrane, Th boundary and
oniy

not

thc

condition ib it is is easy ca.sy one co-

ahplicable applica-bic
to shew tbat a.rc arc

1 undci'
tbo == 10 =

1 thcsc
normal

circumstances circumstfLnccs
functious = or or w = W

is ~h 0 Is 0 ~ln = 0,
whieb involve

ordinale ormate

cos (

7M tt

ces

m
??t

(x

lU ), tthc

orjgui ongm

being omg

a.t a corner

of th

square.

Thus

thc

vibration

22~.]

WHE, ATSTONE )S

FIGURES.

315

thc which

noda.1

systom

dpends applicable

of the is composed ouly on thc symmetry to a square plate.

diagonals. of the normal

two

rcsu~t, is fonctions,

This

strictly

shcwn figure

in Fig. 45. If tlie other sign bc takcn, to tlie other diagonal. with rei'ercnec

wc obtain

a similar

316
Withthcothcrsign

VIBRATIONS

0F

PLATES.

227.

wcobttun

a. systum ruprcscnting inscribcd square. Thcsc

eouiposcd

of th

diagonale

tcgcthcr

witb

thc

to t)tc membrane, aro strictly appHcab)c foDns, which of sand 0)1 a, square obtained p1atu th ngurcs rescinble by mcans of Th squence hve bccn than expcctcd. more might closu)y Frum 176 wc sce that, if /t were toncs is howcvcr quite durrent. from thrcc thc furm bctwec!). interval (4.3) dcrivcd thc zo-o, bc i'rom and two, woutd nodcs, (41) or (42) durived primitive and th interva.1 between (41) or (42) a.ud (4M) or (47) l-4-(i29 octave tbc Wbn.tcvcr of be 2-43.')8 octa-vcs. may bc th value wou!d and tbn tlie same have exacte pitcb, furms (4!) !U)d (42) shouki to tbeHrstWitb respect samc sbould be true of (4(i) a.nd (47). CbLidrit's witb is not in a.grecmetit resuit moitionod pair this a whoc tone, of more than wbo found a dirt'crencc observa.tionH, bc Icft: out of If bowcvcr thc (42) higbcr pitd). (42) giving to thuory Is more thc Aecording satisfactory. account, cumparison fmd (4(i), (47) if (41) gave givc (43) should (~=0), for (43) ~)-, tuund Cbhubu andfor(4G), sbould give~"+. and + respectively. (47) mode of a. square plate bas yet to bc consigravest tbc Tbc nodus in tbis case arc tbc two Hues dra.wn througb dered. That thcre must Le sueh a mode sides. middio points of opposite but of symmctry, from considerations will 1)G shewn prcscntly bas yct nor tbo pitch, tbc fonn of Hic normal function, neither = 0. A rongh case of cveM for tlie particidar beeu dctcrnnucd, of on an as.sumed howcvcr type calcnlatioli mny bc founded 228. Thc
vibration.

228.]

GRAVEST

MODE

OF

SQUARE

PLATE.

317

for axes, thc form !o = a; satisfies couditions propor for a free edge \7*M = 0, as wcll as the boundary th actual This is at ail points of the porimeter corners. cxcept assume if hold at l'est by in fact tlie foi'tn which thc plate wuuid at thc corners four forces uumericidiy acting equa!, pcrpendicuIf wc take tlie nodal lines thosc at tlie ends of eue diagonal larly to tlie plane of thc plate, and those at the ends ofthe other diagonal beh)"' in one direction, tins it follows that w=~cos~~ direction. From in the opposite would werc if thc mode of -vibration, a possible lu tlie four corners. concentratcd equally bc mass plate By (3) 214, we of the

sec that

For

th

kinetic

:ulditionn.I

if p be enorgy, m:tss at eacli corner,

th

volume

density,

and

~)/ thc

whcrc result tends

dnotes

tlie mass

of tlie

to become

accurate

plate whcn~jf

without

the

loa.ds.

This otherwhen we

less than wi.sc by 8f) it is scusibly not very th error is probably jtf=0, should have 2 4 q b~ 2-~ ~=p(l~ a. p'Lc!) which is somewha.t giving next a.fter this is whcn tlie diagona.~ = 0, would Le given by if too

is re~tLvuiy grcut; But even tlie truth. grt. In this

c~e

high. a.re nodes,

gra.vest of which the

Th

mode pitch,

(sec 174).

318
Wo be may conclude

VIBRATIONS

OF

PLATES.

[228.

that

if thc

,-(/themterv~bctwecn than somewlmt greater Chludni makes tho intcrval 2~9 That shortcsb
as

plate wcrc sud. would tones thc two gravest l'SU;. 1.1.. the ratio that -Pd express o by thc t~ Afth. m~tcri~ of thc modes
may

therc
.~cters

must

cxi.t
~-o ncdcs

of vibratiou
bc infcrrcd

in
from

which
su~

thc

two

~Lldcrations

thc

following.

lu Pig. 18.

Fig.

(~)

suppose

that

is a plate i s 6'C

of which

the

Jf~, cdgcs T~~ P' must be our

CO

arc

~or~,

and

th

of cqu~brium, ~t;on modes. uuLhuncnt.1 conceive

over the th~c rcmammg quadrants a distribution of .<; .u-c cqual aud tl~c values that two adjoin, in any such that in th line of each other which arc the inmges at points opposite to thc law vibrato accordmg whulc th If pl~tc of sparation in ordcr to kccp ~ill be rcquircd eoustmiat no detcnnincd, thus th ~hoh plate may be and thereforc ~cd, the lines C~, may be uscd to prove that Thc samc argument free. as rc~rdcd botb the arc nodcs, or in ~!uch thc diagonals ~hich in exi.t modes are togethcr thc diamctcrs just considered and nod~. Ji~onals to other forms aiso be applicd of may symmetry Th principle diameters of nodal iufcr thc possibility thus Wo might of plate tlie Whcn in an ellipse. axes in a eirele or of nodal principal

Fixing of

capable attention

in of vibmting on one of thse,

certam let us

boan~ry (~0), (;)

is a rcgular rcprc'scnt

it i.s e~y hexagon, fonns. pnssib!o

tn sec

that

Fi~.

(4f)),

229.]
It i.s intcrcsting tho form of tlie

PRINCIPLE0F SYMMETRY.
to trace of Chiadni's continuity is graduaDy altered. In the nodal and th thc figures,

319
as

plate whcn thcre are two perpendicular cxa)nplc, mattcr of indiifo'cncc as respects the pitch

circ1e, for it is a. diameters, type of vibra-

As the circlc develops tion, in what into position thcy bo tnken. a. square out tlie position of thse diamctcm corners, by throwing becomes clefiuite. In the two alternatives tho pitch of tlie vibmtinn is dinercnt, for the projccting corners have not t))C Sfunc cfiTIis vibration of a square in cicncy i)i the two cases. plate shcwn to that of a circlc whcn thcrc is ouc circular Fig. (42) corresponds nodc. rite con'cspondcncc of tho graver modes of a hexagon or an cHipsc witli tliose of a cirele may bc traced in likc manner. 230. invariable
mode varies

For

plates shapc, thc

as tlie

of uniform period
square

material the the vibration

and

thiclcness

and

of

of
of

course

ttiat

tlie

comparcd. and assert

conditions boundary fn'e clamped, Whcn th edges that of n~y external the removal whethcr tho inatcrial

in any fondamental linear of dimension, providcd are thc same in aU tl~e cases wc may go further is attcnded bc uni-

hy a risc of pitch, form, or not. Let whethcr

portion and the thickncss

~4~ bo a part of a clamped edgc the rcununder of thc boundary

(it is of no consequence or not), ami be clamped,

the ncw edgc ~173B being also pice ~4C'J3D be remoYed, vibration is sbarpcr TIie pitch of any fuad{nuenta.l cla.mpcd. This is evident, since th altered than beforc tlie change. let thc from the original be obtained system by thc might thc edge ~4-DR The effect introduction of a constra.mt clamping of evcry Is to raise ttio pitch and of thc constt'Mut componcnt, th motion, portion ~IC~Z) being plane and at rest throughout of changes In order to follow thc squence may bc rcmovcd. th from error, it is best to suppose Une with greater security thc of jr' clamping ~1/)/ to For advanee by stages the pitch pxampic, betwcen the two positions ptuto of a ~niform chmpcd vibra.tions

320
in th form and

VIBRATIONS

0F

PLATES.
than circle. that proof an for thc

[230.
inscribed

circle

of a rgulai' is lower hexagon tlian for tlie circurnscribed higher a plate it free, incrcases the a particular It plate, is clear is is not period. case. that thc true In

WIien tlie

addition it

to may

cdgc always to notice be sufncicnt ~7~ Ion.ds at A, C.

of this but to the

is a na.n-owthin

itscifwithoutinertin. addition

cn-rrylng hrcadth

indicated bur, and consideration therc tho

by thc dotted tlierefore ~Ot shews lowcr of th that limit gravest

line thc fur of

would

the stifrncsa of th augment of vibration. Thc same period free by for plate of a sufficicnt givcn area elongation to excd th

a uniform

is no period

pitch

any astiignabic form ofgrn.vest If altered linear virtue The an tlie

be made component may W!ten thc cdges are clamped, quanti ty. the cirele. pitch is doubtless of

a plate, the thickness, including t!tc period in the same proportion, is proportional to as in cvery case of a solid body vibrating dimension, of its own elasticity. dimensions period as th square root inversely be constant, and directiy as the square of th substance. of volume also varies

be th in

modulus, if mass of unit

of Young's root of tlie

with of thin wood whose 231. square plates Experimenting to onc pair of sidcs, W heatstone found thut ran parallcl grain was difforent as the apof th vibrations thc pitch according nodal Unes were paraUel or pcrpendicular straight proximatcly effect dopends on a variation of th wood. This to th fibre in the two directions. Thc two sets of rigidity in tlie cannot hc combincd vibrations djfferent periods having it is not possible to mal~e such and conscquently usual manner, in th flexural of wood vibrato with nodal The a plate diagonals, inequality bc obviatcd th ratio of the of periods hy altcring may however th ordinary mode of superposition nodal sides, and tlien giving is again possible. This was verified diagnnals by Wheatstonc.
'J~.T'r~j'.lHM.

231.]
A furthcr

CYLINDER

OR

RING.

321

to Konig 1, it is only neccssary that th periods that th sides of a rectangular plate the vibration ratio, that (for instance) to one pair thrce nodes new nodal of vibration. 232. W!icu the whose plate thc difficulties of sidcs paralhl figures

of the principle application In order that two modes of

of superposition is duc vibration may combine, Now it is evident agre. may with be taken in such parallel having a caso modes a two nodes

with th vibration may agrcc in pitch to t!)e other of sides. In such pair arise by composition of th two primary

vibrations of tbe

are

to be considered

is naturaUy curvcd, nmch incrcascd. But due a free to curvature edge; aud as being th best of analytical

are gcnerally question thcre is one case ia which thc complication is more than of compcnsated by tho absence this case happens to be of considrable interest, of a bell which at prsent and a uniform admits

reprsentative treatmcnt.

sitell of circular A long cylindrical nesa is evidently of vibrations capable

in winch

section of and the

tluck-

flexural surface

character cylindrical,

th motion

axis

remains

at

rest

'while

of every part is perpendictilar to the generating lines. The problem as one of two dimensions may thus be treated and dpends the consideration of th potential and upon only, kinetic of thc various deformations of which tho section energies is capable. vibrations area about Thc depcnding Tlie same of a ring, an external also analysis formed by th axis. is susceptible of two classes of vibrations and rigidity, of straight by resisting which exbars, th to th corresponding applies revolution of a small closed

th

cylindcr,

or ring,

analogous bars. however, When, small become bcnding tension vibrations than In th flexural is opposed; depcnding those which limiting vibrations as each that case a

on extensibility and flexural rcspectively and to th longitudinal lateral vibrations the in cylinder comparison as in the are is thin, with case the those

forces

and, on bcnding have their of an become whole, of part

graver in origin thin ilifiiiitely

of straight and more longltudina.1 shell (or

important rigidity, thc ring), of tho th supits natural

circumfcrencc position length R.

independent and may the

of any extension be calculated on retains

cii'cumfcrence

throughout

tho

motion. rnRt!)));). 186i, cxxti. p. 238. 21

322
But analogous of depending th although to th on

VIBRATIONS

OF

PLATES.

~333.
to be considercd are

tho tlie

vibrations vibrations to rsistance of

about

transverse

of straight bars in respect we must not fall into nexure, are exctusively of a. cylinder or direction of the of no extension.

t)~at t)iey supposing a motion It is indeed normal. casy to sec that in the each pf).rtictc is displad ring in which with the condition radius wou!d be incompatible common mistake In order to

it is neccssary to aseribe to this condition satisfy a ta.ngentia.1 as wcU !ts a. normal cach pa.rt of the circnmfcrence must di~'errelative whose inagnitudes satisfy a certain motion, Onr nrst stop wi)l be the investigation of this cntial quation. quation. of tlie circlc 'being a, let th equiradius original of the cireumfcrcncc be dcnncd of any clment )Ibrium position the motion let the polar co-ordiDuring angle by the vcctorlal natcs of the cl meut beeomc 233. The ?'=ft+8r, If ds rcprcscnt we have the ~=6+M. curve corresponding to

arc of the deformed

(~)" = (af~)' = (~8r)'+

\vl)cncG ~?-, wc nnd, by negiccting the squares

(~ + f~)'
of the small qnantitles

.(1), .(y,
as the required In time whatcver t, 8r may relation. manner be cxpandcd the circle may be deformed original in the sries theorem by Fourier's cos 2~ + 7?, sin 2~ + at

8r = ft {~1. cos

+ J9, su) <?+.

+j4~cosM~+~sin~+.}.(2),

and

the

condition

corresponding nfno extension

tangcntia! will be

disptaccmcnt

required

by

the

~=-~l,s:n~+73,cos~+.smM0+-"eos?t0tho constant that might be added

?t to 86 being

M. omittcd.

.(3),

233.]
If energy o-<

POTENTIAL

AND

EINETIC

ENERGIES.

323

denote

the

mus

of

th

clement

the

kinetic

T of the whole

motion

will be

th

products

of

the

co-ordinatcs

disappea.nng

ia

tho

intgration. We Lct be havo now to cn.kula.tc of curvaturc of F~wc may on tho materia,! tho form of tho K energy potent!al eletncnt f~, thcn for tho 1\" whero ~Is Thus a

th ra.dms clment

of any take and

coi-responding constant

dcpcuding

~f~(8-j, on the thickncss.

Now

and

for

in the

small

terms

tl)c

distinction

bctwcen

and

<? may

bc

neglected. Hencc

aod

in of~.

\vhich

thc

summation

extcnJs

to

ail

positive

intgral

values

324

VIBRATIONS

OF

PLATES.

[333.

The energy, without

tenn

for which

n = 1 contributes to a. displaccmcnt

as it corresponds dformation. when th

nothing of the

to thc
circle as

potential
a whoc,

is defined as of tlie system configuration above by th eo-ordinates for f7'a.nd V J,, ~t, &c., t]ic expressions coin otlier words, tlicso are tlie ))or)~ involve only squares whose I)armo)ilc 'variation thc ordinates, Independent expresses Wc sec that vibration of tlie system. only thc terms suitably, origiu of
'n 8~=-O

Ifwcconsidcr by taking the

invlving

cos?:~

sin~

wc have

= a.A cos 110, 8r ~=~~cos?~,

r, siii ?td "sin~(7),

(7),

for by Huppe ring in a mcmoir pubHshcd in CrcIIc, Bd. 03,1871. His mcthod, more comptctc though than th preceding, is less simpJe, in consequence of his not rcthat the motion to cxplicitiy eognising contempla.tcd corresponds given complete inextensibility to Chiadni 3' If we rcfcr the ratios cach touc to of thc the circumfcrence. frc(~icnclca 7' thu 0' of thc series, wc Dnd for of the toncs of a ring

This

resuit

was

According arc as

chara.ctcristic 2'778,

of the 5-44.5,

gravcst iuturvaJs 9,

13-4.4, from
cqual

&c. thca.hove thcorctic:Ll

Thc formula?,

corrcHponding by
making 7t

numhcrscbt~iticd
succcssivcly

to 2, 3, 4, are 12-87, A'c.,

2-828, agrccing prctty nc:u'!y

5-423, ~'it.h

8-771, titosc- fonnd

cxpcrimcutaDy.

234.;]
234'. When

POSITION

0F

NODES.

325

= 1, the

TIie principal anticipated. disant and Ims four nodcs, called nodes arc not, however, places a maximum. is ttiere motion tangentiat vibration In M gnerai of t!ic at thse

is zro, a.s might have been frequency mode of vibration to ?! = 2, corresponds from each other soby UO". Thse of absolute In fact rest, for the tlie taugentia).

normal motion. is hali' thc maximum points the maximum motion is for t)ic ?t"' turm tangcntial normal motion, and occurs at the nodes of

maximum

thc lattcr. Whcn a bu!I-s)tapcd body is sounded by a blow, th of maximum of thu blow is a place normal
vibrations, and tlie same is truc

point
motion thc

of

application of thc resutting vibrations room are

when

excitcd

cxperimcnts. howcvcr more casily thc -wctted migci' carried the tlie rcsulting soft part sound of tho

as gcneraHy in Iccturcby a violin-bow, Bu!!s of glass, such as nnger-glasscs, arc iuto j'cgular thrown vibration with by friction round same the circumfcrcncc. chcitcd Ttic pitch of aa of that by a tap with motion tangential

is the

but inasmuch as the finger; th production beU bas been very gonerally of a vibrating ignorcd, bas been fc!t as a difficulty, It is now in tliis manner of sound to point out that the cffect of the friction is in necessary scarccly the first instance to excite and of application of th motion is grcatest, vanishes. 235. The existence tangential friction is the therefore motion, place where and wherc the that thc the point tangential normal motion

of tangential vibration in tlic case of a bell manner. A so-called in thc following rcwas verified. air-pump to a table, opcn end uppermost, cciver was sccureiy fastened and set A small chip in tlie rim, with th molstencd into vibration nnger. the light of a. candie, gave a bright spot whose motion reflecting with a Coddingtou lens suitably could be observed nxcd. As the the hne of vibration was scen to rcwas cai'ricd round, nngcr that an angu!:u' double of the veloeity of excursion the amount of (indicatcd by the length was rinitc in cvery position. variable, though light), in observing th correspondence somc difficulty however, of vibration direction and thc situation thc momcntary volvc with of cxcitoncnt. to apply
To crfeet this satisfactoriiy

and nnger; th line of There was, bctwccn

of the point nocessary It thc'n

it was of one

found point.

th friction

in

the

ncighbourhood

326

VIBRATIONS

OP

PLATES.

[235.

whon thc boll was the spot moved tangentially or 270 degrees and excited at points distant thcrefrom 0, 90,180, points norma.Iiy when tho friction was a-pplied ai the intormediate Carc is somctimes to 45, 135, 225 and 315 dcgrecs. corresponding bccamc vident that required without iu order to ma.ke the bell vibrato in its gravest mode sensible be a.dmixture a smn.U of overtoncs. load at any

of tho c!rcumferencc, point a slight a.ugmcnta.tion of pcriod accordcns~cs, which is different or coincides with a node of the normal ing as the Ioa.ded point of the tangcntiai ca.so than in thc latter motion, being greater in the former. therofore on the Th sound produccd dpends p!a.ce of excitation interfrence give diffurence between uomeuon may often in rise tlie gcncral to beats, both whose tones arc frequency of the two toncH. of largo and hcard, by is equal to the This phchells.

If tliere

frc(~)encies bc obscrvcd in th case

END OF VOL. I.

C.\MDKUXm:

rtUNTKD

!jY

C.

J,

CLAY,

M.A.,

AT

TOE

U!f!Vii:!(H!'[Y

rttHSS.