• Author / Uploaded
• Anton Styles

• Categories
• Flauta
• Naturaleza

Citation preview

Bergeron on Flute-Making Author(s): Michael Wright Source: The Galpin Society Journal, Vol. 29 (May, 1976), pp. 26-34 Published by: Galpin Society Stable URL: http://www.jstor.org/stable/841857 . Accessed: 19/10/2013 19:47 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

.

Galpin Society is collaborating with JSTOR to digitize, preserve and extend access to The Galpin Society Journal.

http://www.jstor.org

This content downloaded from 128.228.173.41 on Sat, 19 Oct 2013 19:47:19 PM All use subject to JSTOR Terms and Conditions

MICHAELWRIGHT

Bergeron

on

Flute-Making

of Manueldu Tourneur by L.-E. Bergeronwas THE first edition publishedin two volumes,in 1792 and 1796 respectively.The

secondedition,alsoin two volumes,appearedin I816, 'revised,corrected, and considerablyenlarged'by P. Hamelin-Bergeron,who describesL.-E.Bergeronas his father-in-lawor step-father.In fact, both names were pseudonymsfor Louis-Georges-Isaac Salivet, an eminentlawyer.1,2 The book is intendedfor amateursandcontains,interalia,detailsof (notexclusivelylathe-work)anda number manyworkshopprocedures of specificexamplesof work.The sectionFlitesetflageolet appearsonly in the secondedition,VolumeII, ChapterII, SectionII, pp. 43-52. The plate, No. XI of Vol. II, reproducedhere (Plate I) on a reducedscale,shows the instrumentsapproximatelyhalf-size.I have no informationon Gallet,the artist,but NicolasLouisRousseau,the engraver,is known to havebeenworkingin Parisaround1815.3The four-keyedflute(fig. I) certainlyseemsto correspondto Parisianflutes of aboutthat period(makingallowancefor the artist'sor engraver's failureto reversethe image),especiallydown to the swivellingEf/D# key; I shouldbe interestedto know of any flute which is found to correspondcloselyto thatshown both in detailsand dimensions. It seemsprobablethat Salivetdid not play the flute, which could explainthe errorsin fingering.In view of the manyminorinconsistencies(to whichI havenot drawnattention)it is questionable whether he madeanyflutes.He may merelyhaveobtaineda descriptionof the work from a flute-makerandhaveincorporated thisin the book with andadditions.Thiswouldexplainwhy he states only minoralterations thatthe keysshouldbe bought,while elsewherein the book he gives ampleinformationon the proceduresinvolvedin makingsuchpieces, and why he describesthe use both of a custom-builtproportional calipersand of a plug-gauge;the latteris quite redundantif one is workingfromtheplatewherethegaugehasto bemadeby measurement with the calipers,but would be the bestand most obviousmethodif one was copyingan existingflute. I havenot beenslavishlyliteralin my translation but I haveavoided the wholesalerecastingof passageswhich would be necessaryto renderthe senseof some passagesin reallygood English.I havegiven 26

This content downloaded from 128.228.173.41 on Sat, 19 Oct 2013 19:47:19 PM All use subject to JSTOR Terms and Conditions

the Frenchfor some words,which seem importantor interesting,in brackets.I havebeenliteralin translating corpsdumilieuandavant-pate becausetheseexpressionsseem to be interestingsurvivalsfrom the daysof thethree-pieceflute.My reasonfor beingliteralovercoupe-vent will be apparent,and I have left ligneandpointuntranslated as there areno Englishequivalentswith exactlythe samemeaning. The unitsof measurementare indeeda problem;by this time the metrewasestablished andthepiedderoiwasno longeranofficiallyvalid as the old units lingeredin common usage the pied was but unit, so that increased slightly 3 piedsusuelsequalledI metre.Thesubdivisions remainedas before: i pied= 12 pouces(not to be confusedwith doigts) 1pouce= 12 lignes 1 ligne= 12points.4 Both the piedde roi and the piedusuelare greaterthan the English foot; 3.280899feet = 3.0784piedsde roi = 3 piedsusuels= i metre.4 We cannotbe sure which systemof units was used in makingthe plate,but probablyit was the laterone. FlutesandFlageolet The transverse flute,fig. I, consistsof four mainparts.The firstA, with the embouchure,is calledthe head.The secondB is the middle joint (corpsdu milieu).It hasfive holes,of which threegive the notes G, A and B naturaland the two others,which are coveredby keys, give B flatandG sharp. The thirdpartC, the upperfoot (avant-pate), hasfourholes,giving D, E andF naturalandthe key for F sharp.5The lastpartC, the foot, hasthe key for E flator D sharp. Thefirstof thesepartsis madein two piecesA, a,joinedby a hollow cylindercalled the slide (pompe)E, used to alter the length of the instrumentto raiseor lowerits pitchasrequired.It alsohasa spherical cap or bung F to end the flute.Betweenthe capand the embouchure is a corkbung which mustbe correctlypositioned. The wood most commonlyusedfor makingflutesis Spanishboxwood, which is preferableto Frenchboxwood. West Indian6woods suchasebonyor grenadillaarealsoused,andgive a purer,clearertone. Choosequartered wood cut with the cleaver,not the saw, becausethe cleaverfollowsexactlythe grainof the wood while the saw doesnot. Similarlydo not choosewood with a twistedor wavy grainwhichis morehandsomebut cannotbe rivenstraight. Choosefourpiecesof equaldensityand,asfaraspossible,of the same 27

This content downloaded from 128.228.173.41 on Sat, 19 Oct 2013 19:47:19 PM All use subject to JSTOR Terms and Conditions

colour and rough out the outside of each between centres leaving the diameter and length greater than the finished size. Then bore through them a cylindrical hole much smaller than that shown in the figure, using a wooden collar-plate. Although the wood should have been sound and well seasonedwhen chosen it is best at this stage to put the pieces in a place shelteredfrom sun and damp to let them settle. When the roughed-out pieces are all perfectly dry, enlarge each hole with appropriatetools held upright in a vice, passing the pieces over them in turn, turning gently by hand and almost without pressing. The tools, called reamers (perces)by instrument makers, are only special taper-bits (louches7)which are made much longer than the piece they are intended for, making them more convenient to use. Their shapeis shown in fig. 3 which depictsa large reamerfor finishing the flute, as describedbelow. The first reamer correspondsto the part AB, the second to CD, the third to EF, and the fourth to GH, extended by about half the length CB. Without this extension a reamer could not enter the workpiece, whose bore is still cylindrical and therefore narrower than it should be. Make these reamers slightly narrower than the correspondingpart of the large reamer with which the bore of the instrument will be finished. These tools must be very sharp. If sharpeningreduces the diameter they are still usable; as they are longer than necessaryat the wide end as well as the other it sufficesto work the tool in up to the appropriate position. Now make the sockets and tenons which connect the four parts of the instrument. The head A, which, as stated above, is in two parts, has three sockets: two for the slide E and one for the tenon of the middlejoint B. The latter has a tenon at each end and so has no socket. The upper foot C has a socket facing the middle joint and a tenon at the other end fitting into the socket in the foot. Mount each piece on a conical mandrel like that shown in fig. 46, plate 17, volume I,8 having the same diameter as the reamer. Leave a bobbin rather wider then the flute joint, on which to place the cord. Spareno pains in making these mandrelsbecauseif they do not run perfectly true the outer surface will not be concentric with the bore and the parts will be of uneven thickness,which will spoil the purity of the sound and prevent the perfect alignment of the different parts of the flute. It is usual to mount the end of each socket with a ring of silver, horn, or ivory. Silver is preferablebecause its purpose is not so much to embellish the flute as to strengthenthe socket which always tends to 2..8

This content downloaded from 128.228.173.41 on Sat, 19 Oct 2013 19:47:19 PM All use subject to JSTOR Terms and Conditions

be split by the forcingin of the tenon which is wound with waxed threadto makeit airtight. After fitting them carefullyfor size, glue thesemountson, as you would thoseof a snuffbox, andleavethem to dry.Thenput the parts backbetweencentresandgive the outsidethe shapeandsizeshownin the figure. Take careto make the diametersof the tenonssuch that they go easilyinto the socketsso thatthey can be wound with waxed thread to makethem fit firmly;makecirculargrooveson the tenon so that the threaddoesnot slip. Then cut the holes,as follows.Begin with the openholes,andfind their positionsaccuratelyby tracinga line on eachjoint parallelto the axisandmarkingon it the centreof eachhole in the positiongiven by the figure.The holescoveredby keysarelikewiseplacedaccording to the figure,but theirpositionon the circumferenceof the flute is not critical;theirdistancesfromthelineon whichtheothersaremarked are adjustedto suitthe lengthsof the keys (whichshouldbe obtained in advance),becausethe end of the key-touchshouldcome to the point shownon the figurefor easeof fingering. As therearenow writtenrulesfor the positionsandspacingof these holes, andeven the bestcraftsmenareonly guided by exemplars,we havechosenone of thebestandhavetakenthe utmostcarein the making of the plate.As canbe seenin the figure,the holesareunequalin size andtheirdiametersareshown on the figurewith equalprecision. However, althoughthe plate may be exact,paperalwaysstretches more or less duringprintingand the effectis uneven,dependingon the qualityof the paper,so we havethoughtit necessaryto drawtwo lines, above and below fig. 2, showingthe exact sizesand spacingof the holesin fractionsof an inch (lignesetpoints).The lower line gives the exactlengthof eachjoint. The upperline givesthe exactdiameter of eachhole andthe distancesbetweencentres. The bore diametersat the bung, under the last key, and at the bottom end, aremarkedon the section,fig. 2. For makingall thesemeasurements with precision,we havedevised the calipers,fig. 4. Thesecalipersconsistof two armsjoinedby a small rivet at A. The two points d, e, are like those of internalcalipers and as they can crossone another,the calipersmay be used for internal and externalmeasurement. But the mainadvan(maftre--a-danser9) of their construction is their tage accuracy. One armcarriesa graduatedsectorB at its upperend,andthe other ends in a point or index, C. As the armsarejoined well below their mid-pointsthe openingBC is much biggerthan de althoughalways 29

This content downloaded from 128.228.173.41 on Sat, 19 Oct 2013 19:47:19 PM All use subject to JSTOR Terms and Conditions

proportionalto it. Consequentlywhenthe openingdeequalsone ligne, BC equalsfour, five, or six, accordingto the proportionsof the caliperarms. Thus if the sectorB is markedoff at intervalseach corresponding to an increaseof openingdeof one ligne, theseintervalscan easilybe subdividedinto four,six, or even twelveparts,thusallowingaccurate of the smallestdimensions. measurement When the positionsof the centresof all the holes are found use springcompassesto tracearoundeach a circleof the given diameter. Then open them all to the samesize, smallerthan the smallesthole, with a pointeddrill.Thenenlargeeachto its properdiameterusinga conicalrose-cutter,fig. 5, withoutin any casecuttingto the line. As a furtherprecautionuse a conicalgauge, fig. 6, markedat the exact mark diameterof eachhole, enlargingeachhole untilthe appropriate the is outside. When flute to the hole but stays completedthe goes up holes are openedto theirexact diameterswith a penknife.This can only be done by tryingthe fluteandby paringaway wood untilthe soundis perfectlypureandclear. If too muchwood hadbeenremovedat firstit wouldnot be possible to put it back,which is why we advisethat the gauge markshould stayoutside. The holesshouldbe undercut,whichis done with a mountedrosecutter(fraisea queue)describedin vol. I, p. 86, andshownin fig. 12, pl. ii of the samevolume.'0 Now put the slide together and assembleall the joints, having woundthe tenonswith waxedthreadas alreadydescribed,so thatthe open holes are all in line. Now finishthe bore by passingthe large reamer,fig. 3, down it. The shapeof this reamershouldbe exactlya truncatedcone cut at right-anglesto its base,with a hollow insideformedas partof a circle not concentricwith the outside,as shownin section,fig. 7. The edgesof the reamershouldbe two straightlines,andshouldcut perfectly.The backshouldbe highly polishedso as not to chafethe wood, becausethe qualityof the instrumentdependsmainlyon the smoothnessof the bore, whose sidesshouldpresentnot the slightest obstacleto the passageof the air. Fix this reameruprightin a vice by its tenon L and slip the flute over it, headfirst.Then, grippingit with both hands,turnit gently, almostwithout pushing,until it reachesthe line A which marksthe exact diameterof the upperopening.As the diameterof the tool is reducedin sharpening, this line approachesthe tenon; thatis why the reamershouldbe madelongerthanseemsnecessary. 30

This content downloaded from 128.228.173.41 on Sat, 19 Oct 2013 19:47:19 PM All use subject to JSTOR Terms and Conditions

Thistool mustbe sharpened with greatcare;nevertouchthe outside but pass a flat whetstonedown the inside, moving it obliquelyall alongthe edge. When making a flute of West Indian6wood such as ebony or grenadilla,the hand may not be strong enough to turn it on the reamer.To assistit, clampto the head of the flute a wooden collar, A, B, one of whichendsin two handles fig. 8, madeof two semicircles like those of a screw-platewhile the other is fastenedto it by bolts C, D, and wing-nutsE. The diameterof the two semicirclesshould be ratherlargerthan the outsidediameterof the flute which they grip; line them insidewith a piece of white leatherto protectthe surfaceof the flute againstbruisingand to preventthe collarfrom slippingwhen turned. Now the fourkeysandthe capat the end of the headmustbe fitted. The latterpiecepresentsno difficulty;it is roughlya hemisphere joined to a tenbn,asshownat F in figs. I, 2, andshouldbe madeof the same materialas the mounts,andcan be decoratedat will. But the method of making keys is entirely foreign to our art. Instrumentmakers themselves havethemmadeby craftsmen who specialize in suchworkso we supposethatthe Amateurwho wishesto makea flutewill buybrass, silver,or gold keys. Firstofferup eachkey to its hole and markthe positionof the saddle,h, fig. I, on the flutewith a fine scriber.Then removea little wood from thisarea,to the thicknessof the plate,and fix the saddlewith two screwsshortenoughnot to reachthe bore. Whenallthekeysarefitted,linetheunderside of eachwitha smallpiece of buffleatherof mediumthickness,trimmedto the shapeof the key. When thinningleather,alwayscut away the fleshside. Positioneach scrapof leatherwith its fleshsideagainstthe righthole, coat the other sidelightlywith fairlythickglue, thenpressthe thumbon the key for a few seconds. Next positionthe cork bung G in the top of the headas shownin the figure.Forthisuse the pusher,fig. 9, whichis merelya boxwood cylinderfor placingthe bung and for removingit for cleaningthe

bore.11

Now the fluteis almostfinishedand shouldgive notes reasonably in tune but a little weak. To make the instrumentmore sonorous, open the boreof the foot from the edge of the hole, using a reamer whose smalland largediametersareshown in fig. 2. Then try all the notesin turnandif any one is not in tuneopen the mouthof the hole whichgivesit with a knife,asmentionedwhen describingthe piercing of the holes. So as not to interruptthe sequenceof operationswe supposedthat 3'

This content downloaded from 128.228.173.41 on Sat, 19 Oct 2013 19:47:19 PM All use subject to JSTOR Terms and Conditions

the Amateur would have obtained a ready-made slide, but as this part presentsno great difficulty we think we should describehere the method of making it. This slide is just a tube of silver or brass whose inside matches the bore exactly. That is why it must be fitted before the final reaming. To make the tube, cut a parallelogram of metal whose length is given at E, fig. 2. The width should be such that in bringing the edges together we obtain a cylinder one ligne less in diameter than the finished slide. Silver-solder the two edges together by the method given on page 201, vol. I. De-scale and thoroughly clear the inside with pumice. Then slip it onto a conical steel triblet, fig. Io, of exactly the same diameter as the reamer at C. The part ab of the triblet is exactly the right shape for the inside of the slide. Hammer very lightly over the whole surfaceof the tube until it comes down to a. The bore is then perfectly true and equal to that of the flute. Chuck the piece on a cylinder of hard wood in the lathe, trim it to the length given in the figure and turn the inside'2truly cylindrical. The slides of some flutes are made of two tubes of brass or silver, one of which slides over the other and fits within the thicknessof the wood. This arrangementis perhaps more pleasing to the eye but it does not make the flute better, and as it is harderto make than the one describedwe have thought it best to keep to the latter. Fig. 12 representsan old-fashioned flute. This has no slide like the above, but is equipped with two alternativejoints (corpsde rechange), figs. 13 and 14, which serve to shorten the instrument more or less by substituting them for the middle joint. The one shown fitted gives the greatest possible length. Otherwise, the flute is made as described above, only the reamer must be passedthree times to fit the alternative joints to the other parts. Notice that this flute has only one key for D;13 the sharpsand flats which were obtained by keys on the other flute are here made by des doigts). cross-fingering (croisement The piccolo (octavedeflate), fig. Is, is merely flute no. I reduced to half-size. This has only one key like the flute in fig. 12 but has a slide like that in fig. I. The flageolet, fig. 16, is made in two parts, the head A carrying the beak B and the window (lumiere)with the wind-cutter (coupe-vent) C, and the body D with six holes, four above for the notes D, F, G, A, and two below (shown as dotted circles) for E and B. These two parts are first roughed out and bored separatelylike the partsof a flute; then they are joined and finished outside and inside, using for the latter 32

This content downloaded from 128.228.173.41 on Sat, 19 Oct 2013 19:47:19 PM All use subject to JSTOR Terms and Conditions

:

if

IN

I

N

'44

Asg

This content downloaded from 128.228.173.41 on Sat, 19 Oct 2013 19:47:19 PM All use subject to JSTOR Terms and Conditions

N

operation the reamer,fig. 17, which differs from that in fig. 8 only in size. The holes are cut in the positions given on the line below the figure. Their exact diametersare given on the same line; notice that they are all equal except the middle one which gives G. Now the lower part, the body, is finished and it remains to fashion the top of the head and the window; proceed as follows. Turn a cylindrical, or rather very slightly tapered, bung of wood whose colour contrastswith that of the instrument, of the length ab, fig. 16. Try it in the head and fit it for size so as to be airtight. Then withdraw it and make a slot inside the flageolet in the part where it fitted. The slot, D in fig. I8,'4 does not go beyond the tangent to the circularbore and extends a little below b, fig. 16, the end of the bung. Make a flat along the bung of the same width as the slot. Before putting it back in and fixing it, it is necessaryto make the window and windcutter C. Make two very small holes with an awl working from the outside of the instrumentat b, that is, a little above the end of the slot. Then make a transverseoblong opening at this point, perpendicular to the outer surface, of the size shown in the figure. For this use a well-sharpenedgraver whose edge is precisely the same width as the opening should be. The same tool is used to make the bevel called the wind-cutter which adjoinsthis opening. This bevel faces the middle of the channel formed between the slot and the flat on the bung, and so divides the stream of air from the channel. (Hence its name, 'wind-cutter'.) Throughout this operation it is necessary to use the graver with great care and to cut away very little wood at a time, for fear of causing splits and damaging the edge between the bevel and the bottom of the slot, because the neatnessof this edge greatly influences the purity of the instrument'stone. Then glue the bung in place, making the flat along it correspondto the slot in the bore, to form the channel through which air, having been divided by the edge of the wind-cutter, enters the body of the instrument. When the glue is really dry, carve the undersideof the beak, removing part of the outside of the instrumentand much of the width of the bung, using files and rasps.As far as possible follow the shape given in the figure. We say 'as far as possible' because this curve is not strictly the only one permissible;the point is that one should be able to grip the end of the beak between the lips. Flageolets are often made of ebony, grenadilla, or other precious wood, and it is usual to garnish the beak, the end, and the sockets 33

This content downloaded from 128.228.173.41 on Sat, 19 Oct 2013 19:47:19 PM All use subject to JSTOR Terms and Conditions

joiningthetwo parts,withivory.To do this,roughout the outside of eachpieceseparately, stickon ringsof ivoryof sufficient lengthand thickness in thepositionsshown,andturnornamental on mouldings themasyoupleasewhenfinishingtheflageolet. of thebeakis carvedin thesameway,withthesame Theunderside filesandrasps,asif it werenotcasedin ivory. NOTES I CatalogueGinlraldes LivresImprimisde la Bibliothhque Nationale,Paris, I94I; art.Salivet. 2 NouvelleBiographie GinCale,ed. FirminDidotFreres,Hoeffer,Paris,1864; art.Salivet. des Peintres,Sculpteurs, 3 E. B&nezit,Dictionnaire critiqueet documentaire Dessinateurs et Graveurs, LibrarieGriind,196o; artRousseau. Universeldes Poidset Mesures..., Paris,1840, 4 Doursther,Dictionnaire arts.Motre,Pouce,Ligne,Point. 5 Sic.:openhole gives Fo, key F?. 6 Boisdes les: this epithetusuallymeans WestIndianratherthan merely foreign,but (black)ebonycomesfromMauritius,the E. IndiesandAfrica;see C. Holtzapffel,Turning andMechanical London,1843,vol. I, p. 83. Manipulation, 7 Louche:lit. ladle,but this is not what is known in Englishas a spoon-bit. 8 Redrawnhere as fig. i; the lathe envisagedhad fixed centresand the work was drivenby a cordwrappedroundthe bobbin.

FIG. I

thefeetareturnedoutwards. 9 Socalledbecause hereasfig.2; thedetachable cuttingheadcanbesliddownthe Io Redrawn theholein theflute,intoit, bore,andthenthehandlecanbeinserted, through to turnit. FIG. 2

i1 The pusherappearsto be markedso that, when the bung is correctly placedandthe pusherplacedagainstit frombelow, the markisjust visibleat the upperedge of the embouchurehole. Forthe correctpositionof the bung see A. Baines,Woodwind andtheirHistory,London,1957, 53. Instruments p. 12 Sic.:outside mustbe intended. or Sic.: is meant. 13 D# E7shows a 14 This drawing flageoletwhose windwayis formedentirelyby cuttingawaythe bung. 34

This content downloaded from 128.228.173.41 on Sat, 19 Oct 2013 19:47:19 PM All use subject to JSTOR Terms and Conditions