The Aircraft Engineer April 24, 1931

7/27/2019 The Aircraft Engineer April 24, 1931

1/8

Supplement to FLIGHT

ENGINEERINGSECTIONEdited by C. M. POULSEN

April 24, 1931CONTENTS

PAGXA Graphical Method of Stressing Aeroplane Spars. By D. Williams,B.Sc, AJVIJJVIech.E 25Shock Absorbers for Aircraft Landing Gear. By W. S. Hollyhock ... 27In the Drawing Office... ... ... ... ... ... 31

A GRAPHICAL METHOD OF STRESSING AEROPLANESPARS.

By D. WILLIAMS, B .SC. , A.M.I.Mech.E.(Concluded from p.21)

I I . Change of Moment of Inertia of SectionIn case I it wasseen tha t the locus line wasshifted parallelto itself at each boundary line. A change of moment ofinertia I, which results in a change in ;x, produces no suchshift, but merely has the effect of rotat ing the locus linethrough an angle depending on the magnitude of the changein \x. This will be illustrated by taking a main bay AB,in which twochanges of I occur, as shown inTable II.Referring to Fig. 4, draw the positive andnegative sectorswith angles a1; a2 and a3. Assume OmApositive, as usual.Draw the perpendicular at m^ to give the first locus line W2

T A B L E II.

BASSUMED T H A T

ex CC , I a,- ZERO |

an d the points fx and lv With /, as pivot ttt isrotated tothe position ll2 by taking any point y on Ult dropping theperpendicular yy1 on f-f, finding a point y2, such thaty&Jyyi = V-Jv-v a n d finally joining yjv The lat terproduced cuts the next boundary line in gx, and the baseline BJB in nx. At gx ll2 is rotated in a similar way (i.e.,z2z1/zz1 = !X2/n.3) to give the last locus line lls, whichcuts BXB in n2 and makes an angle G with it. This angleis measured. This time the movements of the locus line,as recorded by its point of intersection with BXB, are not


2/8

SUPPLEMENT TOFL IGHT APRIL 24, 1931THE AIRCRAFT ENGINEERmeasured by the length of the intercepts l1n1 and n-pi^, but0nx 0w.2

On, On8 , _ \B I

by th e ra tios - , - an d our first equat ion is

sec ^ ~* + mB) tan 6 = Sfl s and, as usual, the second equation is

p . B gj j = M B - M A + R BAB(In t his pa rtic ular case, since w is zero, an d th ere a re noconcentrated loads, R B = o).The value of *'B is thus found in terms of MA an d M B - Ba yCB would be treated similarly to get another expression forin , and the standard procedure followed.MA and MH, having been found by the abov e met hod, theactual B.M. diagram is drawn as follows :Referring to Fig. 5, draw the sectors and boundary lines

as for Fig . 4. Suppose both rA and mB are negativ e. Layoff OraA and OwH in the negative directions of AA1 and BB1,and draw perpendiculars to give llx and U3(Y) respectively.IIx cuts the first boundary line/ 1 / i n fx and B JB in lx. Insertthe points nx an d n2 usin g the ratios - a nd , as foundin Fig. 4. Join t fxnx to give llit which cuts the next boundaryline gxg in g,. Join g1n2 to give U3(2). The latter cuts 11,(1) inX, . Draw X a X 2 perpendicular to gxg, cutting ll2 in X2 .Draw X,X! perpendicular to/ ' /cutting llx in X,.0 X 1 ( OX2 and OXS are the diameters for the cuttingarcs to be drawn in sub-sectors 1, 2 and 3, respectively.Since w is zero, the loading arcs shri nk into the p oint 0 .The bending mome nts are shown shaded. In this case allthe cutting arcs are below their respective loading arcs,which have shrunk into the point 0, and, therefore, thebending moment is everywhere negative.

I I I . Combination of I and II.Suppose a change of moment of inertia, a change ofdistributed load w and a concentrated load all occur togetherat a poin t in the main bay. Such a point will be a bo und aryline between two sub-seotors in the polar diagram, and thethree types of discontinuity should be dealt with in theorder just given, i.e., the locus line is first rotated to accountfor change in I, then shifted along the boundary line, butparallel to its rotated position to account for the change ofw, and finally moved along a perpendicular to th e bo unda ryl ine , still parallel to th e rot ate d position to acco unt for theconc entr ated load. All such moveme nts of the locus line arerecorded as points of intersection with the base line B1B,and, as before, movements of the locus line parallel to itselfare measured by the lengths of the intercepts on B JB, whilerotations are measured as ratios.A single example applied to a bay AB will suffice to explainthe procedure. The various quantitie s are tabu late d inTable III.Draw the sector and boundary lines and assume 0w Aositive as before (it is convenient to make this assumption

Table III.

always even if mA is known to be negative). Erect a per-pendicular at mA to give 11 x the first locus line which cutsB XB in Zx and Z1/ in /, . Rotate at fx to the position / ^bu t note that f1n1 is not the locus line 11% because the move-ments /j/a and/2 /3 which account for the change of w and theconcentrated load respectively, have not yet been made. Aline through /3, however, parallel to f1n1 gives ZZ2, the secondlocus line, which cuts the next boundary line g'g in


3/8

tlA P R I L 24, 1931 THE AIRCRAFT ENGINEER SUPPLEMENT TOF L I G H T

U1 (i.e., fiti) in y2. Drop a perpendicular y^y1 o n / 1 / to cutWx in X^ OX j, OX 2, and OXS are the three diameters for thecircular arcs to be drawn in sub-sectors 1, 2 and 3 respectively.Draw these cutting arcs and then draw the loading arcs kjcj,k2kz, and s&s as described in Case I. The radial intercep tbetween the two arcs gives the bending moment for anyangular position of the radius.It will be seen from the above construction that the firststep is to draw the locus lines and their rotated positions, andso locate the X for the last sub-sector. Havin g found theposition of X 8 on ZZa(2) we can only obtain the position ofX s on II 2 by first using the ro tate d position of the lat ter. IfU t had not been rotated, U3(2) would have been parallel to it,and X 2 would hav e been found directly by drawing a linethrough X, parallel to g3gr of Fig. 6. The same thing happe nswhen proceeding from X 2 to find X x ; it is first necessary tofind the auxiliary point y2 on the rotate d position of II

before the perpendicular /2X! can be dropped on theboundary line J 1 / to give X x .The above description may appear a little involved, butonce the procedure is thoroughly mastered it is surprisinghow quickly a particula r problem can be solved. I t isimportant to use as large a scale as can conveniently behandled and to make all measurements with care.Note.In each of the above cases a figure has first beendrawn with the object of obtaining the necessary data for thetwo equations and the positions of the locus lines, etc. Theactual bending moment diagram has then, for the sake ofclearness, been draw n separ ately. I t is now pointed out th atit may often be more convenient not to draw a separate dia-gram, but to make use of the first figure for the purpose ofdrawing in the actual bending moment diagram.

SHOCK ABSOR BERS FOR AIRCRAFT LANDING G EAR.By W. S. HOLLYHOCK.

THE mos t generally used shock absorbing media for aircraftlanding gear are oil, air, rubber an d springs. These areemployed in a variety of ways, the two most common ofwhich are combinations of oil-and-air, and oil-and-rubber.The all-rubber shock-absorbing unit is practically obsolete,and for various reasons, springs are n.ot viewed with anygreat favour hi this country, so it is not proposed to discusseither of these latter methods in detail in this article.In both the oleo-pneumatic and oleo-rubber combinations,the oil serves the dual purpose of absorbing the shock ofimpact on landing, and of dissipating the energy so absorbed.The air and rubber on the other hand, give the necessary" cushioning " effect in taxy ing, and in certain cases alsoserve as factors of safety in the event of unduly h arsh la ndings.Oil being incompressible, cannot store energ}7, and thereforecanno t be used alone. Conversely, air and rubbe r beingincapable of dissipating energy , need to be used in conjunctionwith oil to prevent bouncing of the aircraft when taxyiog,and to keep down the size and weight of the shock-absorbingunit .Incidentally, it is chiefly in this respect that springs arenot considered satisfactory, because a damping medium mustbe used in conjunction w ith them , and if oil is used th e w eightbecomes prohibitive. The alter nativ e is some sort of Mo-tional device with its concomitant evils of wear and unrelia-bility. The oleo-pneumatic combination is undoubtedlythe best arrang eme nt in every way . It is the lightest, offersthe least aerodynamic resistance, and is more durable thanothers.The big bogey of the pneumatic leg is the high air pressurenecessitated. A t least, it is considered an objection, thoughit is difficult to und ersta nd jus t why. Certainly accidentshave happened, but aeroplanes have been known to breakup in the airand the aircraft industry has survived theshock. A ctually, there is no reason why pneum atic cylindersshould be any more prone to bursting than engine cylinders,and as regards accidents on the ground, these can be elimi-nate d by rendering th e unit foolproofnot a very difficult

proposition. There is one other disadvantage about thisarrang em ent, namely, the filling of the leg wit h air. Thisoperation, if carried out by hand (as is usually the case)takes something like half an hour to perform. Neve rtheless,to allow such a trivial matter to bias design would show adistinct lack of sense of proportion. As a m atte r of fact,if the unit is properly designed, the leakage will not be greatso that frequent replenishing will not be necessary.The principle on which all oleo combination units are basedis the converse of the hydraulic ram acting either in advanceof, o r simultaneously with , an energy storage device. Theunit consists of a cylinder containing oil and a piston whichon impact forces the oil through a valve or small orifice.In the case of the oleo rubb er arrang eme nt the piston com-presses the rubber after exhausting the cylinder of oil. In th epneumatic leg, the compression of the air may take placeconcurrently with the evacuation of the oil cylinder, or followit, according to the type of leg concerned. I t follows tha tsuch energy as is absorbed by the oil is actually dissipated ;while that absorbed by the rubber or air is stored and givenout again in the form of a rebound action. In th is connec tion,it should be noted th at the rebound action should be controlledby suitable stops and /or thro ttling of the oil passage to pre ventthe aircraft bouncing undu ly when taxy ing. To this end,the rubber is initially compressed from one third to onehalf " g." With air, the higher figure should be taken(owing to its greater elasticity) if th e two actions are consecu-tive as in the case of the rubber leg. Wh ere the oil and airact concurrently, the air must, of course, be compressedinitially to one " g " to ensure correct functioning of the oilvalve. In all cases the retu rn velocity of the oil mu st becontrolled by throttling, though the amount of retardationis not of great importance except where the oil and airact concurrentlyin which case, great care is necessary onaccount of the high air pressure acting on the piston.With regard to the design of the oil valvewhich, by theway, is usually of the needle varietythe cushioning effectof the tyres should not be taken into account. The am ountof energy absorbed by the t yres , even when fully inflated,is not great and when they are deflateda not unknown,occurrenceobviously does not exist. So th at if the leg isdesigned to allow for tyre effect, the aircraft will sustain aserious shock when landing on flat tyre s. WThereas theeffect of landing on fully inflated tyres with oleos not designedto take account of tyre resilience will be to ease, somewhat,the first shock of impact, though there must, of course, be aslight harshness when the tyres cease to deflect.With regard to the actual design of compression rubbersfor use in oleo-rubber units, it is not proposed to go intodetails in this article, as the m atte r has alread y been discussedin these columns. So long aB the pro portions of the individualrubber s are not freakish there is not m uch to choose betweendifferent shapes. The one really im por tan t requirem ent,of course, is a first quality m ateria l. Eve n so, they need to becarefully watched in service as they rapidly deteriorate-particularly under conditions of extreme temperatures.With regard to oleo-pneumatic combinations, there aresome important facts which should be borne hi mind whendesigning such un its. Fir stly, it is absolutely impera tivethat the leg be so arranged as to be quite foolproof as far asthe releasing of the air pressure is concerned. I n other word s,it must be so protected by locking devices, that it will notbe possible for a mechanic to release any pa rt in such a m anne rthat it can become detached from the leg while there is anyappreciable air pressure behind itwhether the leg is actuallyin place on an aircraft or not. The possible consequence ofsuch an accident need not be elaborated if it is rememberedth at t he air pressure m ay be of the order of 1,000 lb . per sq. in.Secondly, owing to the high pressures attained, joints andglands need careful consideration in order to eliminate lossof pressure as far as possible. To this end, glands exposedto high air pressures should be so arranged that oil is presentto keep them tig ht. In fact, it is really desirable to avoidglands exposed to air at high pressure, altogether, and it isquite possible to do so if the right arrangement is chosen.Joints which are exposed to high air pressure should bebalanced by high oil pressure on the reverse side, where


4/8

SUPPLEMENT TOFLIGHT TH E AIRCRAFT ENGINEER APRIL 24, 1931possible, for the same reason. Where this is not feasiblethey should be thoroughly sealed, even at the expense of alittle extra weight. In such cases it is a good idea to weldthe joint if possible or even to go to the extent of machiningout of the solid and thus avoiding a joint altogether, the lattermethod ensuring both absolute tightness and accuracy.Another point which might conceivably be overlooked, is thefact that in certain arrangements the piston rod decreasesthe effective volume of the oil cylinder and if the oil has topass from one side of the piston to the other it may be neces-sary to extend the rod beyond the piston so that the volumeon either side is unaffected by the movement of the rod.The following example indi-cates the lines on which toproceed in the design of anoleo-pneumatic leg; and as th


5/8

24 , 1931 THE AIRCRAFT ENGINEER SCPPLXMXMT TOFLIGHTActually, the initial oil pressure is slightly greater than atthe end of the stroke, but the difference is negligible when areasonable factor of safety is allowed.

Air Volumes and Pressures.If the static load on the tail skid is 10 per cent, of theweight of the aircraft, the static load on one wheel will be4,500 X 8-284,500 lb. Static load on oleo is, therefore,

= 4,6501b.LetPai = Air pressure with leg fully extended (F.E.)Pa2 = in standing position whilestationary.Pas = t with leg fully compressed (F.C.) assuminga static condition.

P4 = ,i with leg fully compressed in action.Pas = ,, with leg in standing position duringaction.Vs = Air volume with leg fully extended.V, = in standing position.V3 = fully compressed.(See Fig. 2.)

Then Pal Vl = Pa8 Va = Pa3 V3 (Isothermalcompression)and pal Vtn = pai Vs = p,A V3 (Adiabaticcompression)4,650p at = - - = 600 lb. p.s.i.At F.C., since velocity is zero, pa ( = pressure difference Po Pa)= 0.Also, P a i Y1 = pai V2 = Pai (Vx Agt) where t is thetravel of the leg from F.E. to standing position (S.T.) andPai Vi" = pai V3 = pai (Vj Ag T) where T is the totaltravel of the leg.

Hence, pal = paiTherefore, 'Pai

\ V,'Vt - A

Pa i V ,f - AgT\

V ,or

Letand ^ I -

?*!(i - ;-?:) = (i _Pai\

V ,PatPat

= z, n= 1-3= y

Then values ofa^and y ea n beplotted forvarious values of Vlfand where the twocurves intersect will be the true value of0-300-2B0-260- 240-220-20

018'< & 0-16. X 0 - 1 *fe0 - 12^ 0-10$ 0 - 0 8

0- 060-040-02

TRUFVMUeOF l{ /S 86-S cumcfies

y

, "

U R

/L\\'ES1

/

/

k\

3 4TRAVEL (INCHES) -> 8HOUrHOCA

Vx for the particular case under consideration (see Fig. 3).With regard to the ratio , this will be fixed automatically incases where the oiTand air function consecutively ; but wherethe actions are concurrentas in this exampleit must bedetermined arbitrarily. The factors to bear in mind infixing it are that the greater it is (i) the smaller the airchamber and, consequently, the lighter the unit and (ii) theharsher the action in taxying. So that, as is usual in aircraftdesign, a compromise must be made. Generally speaking, avalue of 0-5 is sound and this figure will be taken in thisexample.All pressures and volumes can now be deduced as follows :Vj = volume at F.E. = 86-5 cubic in.V2 = S.T. = V , - A ^ = 86- 5- 31 =55-5 cubic in.V3 = F.C. = Vx- AS T= 86-5-62 =24-5 cubic in.

i T 1 _ , Pa%Vt 600x55-5p ai = pressure at F.E. = = =

S.T. (Isothermal) =

/VA 13(Adiabatic) = pai ( ) =

385 1b. per sq. in.4,650 =7-75

600 lb. per sq. in.Pa s

Pas, =

385 = 686 1b. per sq. in.\ 5 5 - 5 ,p, n ,, F.C. (Isothermal) = * =V,

600 X 55-5 n 1L = 1,358 Jb. per sq. in.24-5Pat = (Adiabatic) = 2,000 lb. per sq. in.

The Oil Valve.The frictional resistance of the oil in passing through the

valve and also the friction of the mechanical parts, will beneglected. It is not possible to calculate these and, in any362*


6/8

SUPPLEMENT TOFLIGHT THE AIRCRAFT ENGINEER A P R I L 24 , 193Acase, they will be partially nullified by the fact that thecompression of the air is not truly adiabaticsince some heatmust inevitably be dissipatedand the air pressure willconsequently be lower tha n calculated. The ultimate error,therefore, is not of great magnitude.Applying the usual hydraulic principles, the velocity-pressure equation may be writte n :

62-4S %j \Ao!WherePd = pressure difference = p, , p< ,p 0 = oil pressure (lb. per sq. ft.)p a = a ir pressure (lb. per sq. ft.)s = Specific gravity of the oil used = 0-95 (say)V = Velocity of compression of oleo = vertical velocityof aircraft x ratio of oleo travel to wheel travel(ft. per second)A,, = Nett area of piston (sq. ft.)and A = Area of orifice (sq. ft.)

V* ASHence Ao* = X X 62-4 X 0-95 sq . f t .2g Pdo r V*1 = - x X 62-4 X 0-95 if An and A,.2g " pa " 14 4are in Bq. in. and pa is in lb. per eq . in.

VTherefore, Ao = 0-584 - 7 =VP< iAllowing for a coefficient of discharge of 0 8, this equationVbecomes An = 0-731 7=-\/pd

10

FIG 5.VELOCITY CURVE

\

\

2 3 4 5TRAVEL (INCHES)

Incidentally, it should be borne in mind that, in order toob tain a high va lue of co-efficient of discharge , the orifice mu sthave a good entry a t each end, and must not be ei ther u ndulylong or excessively short.It is now necessary to determ ine the va lues of V and pa .Since the deceleration is constant, V a = u% -\- 2fs oat F.C.M' 10 x 8Hence / = - = - a M _ ^ = - 70 ft. per sec.22s 8-28 x 2TTherefore, the velocity at any instant is determinable.

The load on the oleo is constant since the resistance isconstant .Hen ce, load = 15,500 = poA? + paAh

00 ~ PaAhTherefore, p0 = 2,120 - 00 60 4 Pn= Po-pu = 2,120 - 1-0604pa

Having obtained these equations, it is now possible tofind the needle diameter at any pointIn p lotting the necessary curves, interme diate po ints at say,a quarter and three quarters of the travel should be inter-polated in order to obtain th e curves more accu rately and alsoto check the other va lues (see Figs. 5 and 6).It is also as well to plot the pressure curves as a crosscheck on the calculated values of pa (see Fig . 4).The values of the various factors should be tabulated asfollows (th e sequence of the columns indic ating t he orde r ofworking out):

Position.

F .E .STF.C.

il9-678-386-844-840

38 549 768 61,0452,000

re p.s

o < .o a2,0972,0902,0792,0572,000

P.;. sq. in.

1,7121,5981,3931,0120

0-1700-1530 1340-1110

0-2710-2880-3070-3300-441

o-S

0-5840-6050-6250-64H0-750

It will be noticed that the curves of "A o " a nd " needlediam eter " (Fig 6) show a sudden change of slope towards theend of the stroke. This is due to allowance having been made

0-5

0-4

o-i

FIG.6ORIFICE & NEEDLE CURVES

3 4 5TRAVEL (INCHES)in the calculations for co-efficient of discharge, which must,of course, be eliminated tow ard s'the e nd. Also, the needlediameter at the end is given as 0 75 in., but in practice theorifice can never be'zer o or the leg would jam. So th at asma ll orifice taust be left and its dimen sions will dep end on th eaccuracy of workmanship obtainable, and the concentricitywhich may be expected to obtain between the needle and thepiston under working conditions. In an y case, a mechanicalstop mu st be provided a t the end of the stroke for sa fety.

Return Stroke.If W = th e externa l force act ing on th e oleo {i.e. suchproportion of the und ercarriage w eight as comes on it),


7/8

24, 1931 SITHE AIRCRAFT ENGINEER SUPPLEMENT TOFLIGHTPo rPa rAo rA hr

Oil pressure during return stroke,AirOrifice area ,, Orifice hole area during return stroke.

an- A,,,At and A;, = Piston and orifice hole areas as before,thi a Vor'Af = W + ParA/i, . ,. / W + p-Ah\therefore, por b e i n g n e g a t i v e , p,,r = ( r, W + Pu rA/,and Pdr = Par = Por = Par T A ,

W ^ = 2 7 - 4 + 1-0604/A ,/The value of W will generally be approximately constant,but if it varies greatly, the calculations should be adjustedaccordingly. For the purpose of this example, W will be takenas constant and equal to 200 lb.If the hysteresis effect, due to mechanical friction and the

inevitable slight loss of heat, is neglected, par may be taken asequal to p,, and p,ircan easily be found for any position.The velocity of return will be given (as for compressionstroke) by the equation VAor =0-731 - 7 =V P d rand, if the area of the orifice is unchanged for the returnstroke (i.e. if Aor = Ao), the velocity can be found directly.In practice, however, it has been found desirable to decreasethe return velocity in order to eliminate sloppiness and con-sequent risk of damage to wing tips. The actual maximumvalue is a matter of individual choice ; but, for this example,will be taken as 4 ft. per sec.To this end, the orifice must be restricted mechanically bysome automatically functioning sliding valve or other suit-able device. It is of the utmost importance, however, thatwhatever method is employed, there shall be no possibilityof jamming in action.Since the diameter of the needle is greatest at the com-mencement of the return strokeit having been fixed for thecompression strokeit follows that even though an auxiliary

2250

2000

toa.sPRESSURE

(

soo

n

v_\ V

TR

fcir

A V E U l ^

FIGC U R V E S OFTOR R E T U R N

4CHES)" 0 1 2 3 4 5 i100

-200'po r

7.PRESSURESTROKE:.

\ 1

valve is used for the return stroke, the orifice must increaestowards the finish. Therefore, the velocity will increase, sothat it becomes necessary to make the auxiliary valve func-tion on an extension of the needle for the last part of the stroke,in order to bring the velocity down to zero. (Care must betaken, however, to ensure that this does not in any way inter-fere with the action of the main valve during the compressionstroke). Consequently, the velocity will increase until thisstage is reached and then fall off comparatively rapidlyanother reason why the maximum velocity should be limited ;as , otherwise, the deceleration of the moving parts will be toorapid.If the change is made to occur at, say, 2 in. from the finish,then the velocity at that point is the maximum (viz. 4 ft. persec) , and from the pressure curves (Fig. 7) pjr is found tobe 554 lb. per sq. in.

4Therefore Anr = 0-731 X 7 = = 0-124sq. in.\/554and Ahr = An + Au, =- 0-288 -f 0-124 = 0-412 sq. in.Since A/,r is constant, at any point where the standardneedle functions Aor = A/,r An ; so that A,,r and V can befound for any position prior to that at which the needle exten-sion comes into action. Beyond that point, the velocity andorifice must tail off to nothingor as near nothing as prac-tical considerations of manufacture will allow.

45>3t

>

0

>

/

15 1

~

FT

_ -

Aor^-1^ ^

V

*>

FIG.8.CURVES Of VE1OCITY& O R I FI C E A R E AR RE TURN S TRO K E

V\

\\

\

0 130-120-110-100-09 ~0 08 Z0 07 &!0-06 fe0-050-0+

0-020-01n? 4 5 6 7 8TRAVEL (1 NCrtES) w w r

The values obtained should be tabulated as under andplotted for checking purposes as before (see Figs. 7 and 8).

Position.

P.CST .F. E

Travel(inches).

02468

ft:p.s. i.2,0001,0456 864 973 8 5

ft:p.s. i.2,1521,13776 5sr.44 35

?p.s. i.

- 152- 92 6B- 57- 50

Vft./Bee.

03-793 954 000

A .,sq . In.

00-0820 1050-1240

In conclusion, it may be well to mention that the unitshould be so designed that the oil level is well above the orifice(with a reasonable allowance for leakage) when inclined at themaximum angle which can obtain in service.

IN THE DRAWING OFFICESETTING OUT TAPER RIBS.

By R. HALEY.The application of shipbuilding formulae to aircraft isfairly easy when dealing with floats and seaplanes ; in fact,it is essential. But to apply any shipbuilding formula towing construction requires a stretch of imagination. Never-theless, the writer will endeavour to show how this can beapplied in the case of taper ribs at wing tip.


8/8

SUPPLEMENT TOFLIGHT32

THE AIRCRAFT ENGINEERAPR IL 24, 1931

+

F/G.JF.

/?/& O/?0MTS.f r

Si,/7^3Y5?/ 7

X

i'7

y

51

05001

%/4 it:V

A \K5 0 L aIfoin121

aa )25-3

a

_K.

(5(!S?W-42/5

/V 7

Ml-51*S&HiHi

e/fl

* - / a - 1 k "

The idea was taken from a treatise on yacht's lines, wherethe author of the treatise had based his assumption on thefact that a yacht's lines are of parabolic form, and he usedA/fc2the well-known formula for a parabola a = . All the

offsets for the lines were calculated and issued to the mouldloft without a single lines drawing having been issued. Theadvantage gained is not the subject of this article.In a " wash ou t" wing tip, the top and bottom surfacesAh?usually conform to a parabola, and the formula a = -

could be altered to 3uit the working out of the ordinates forthe taper ribs by expressing same thus :/ *a =

The writer had occasion to require the true shape of asection of the wing tip at " XX ," known as a " cant " inshipbuilding, and although the layout for the taper ribswas correct in profile, it was found on " lifting " the " cant,"that the new shape was 0-1 inch out in height. Thh was,no doubt, due to the fact that the ribs were not faired in allthree views.As the new rib was in metal, it was necessary that thisshould be correct in profile, and by applying the formulaalready mentioned, it was possible to get the correct outlinefor the taper ribs, and hence the new rib at " XX," Fig. IV.Fig. I. is a profile of the last normal rib marked L.N.R.in Fig. IV.In the table of ordinates, Fig. Ill, the ordinatee for theL.N.R. are marked X and Y, being top and bottom ordinateBrespectively, above datum. Let d equal the diameter of thewing tip tube, K the distance of L.N.R. from wing tip tube,

centre to centre, A = , Fig. II , and k thedistance from L.N.R. to taper ribs, 1, 2 and 3.

x and y are top and bottom ordinates of taper ribs abovedatum, see Fig. II. x = X a and y = Y + a.In the example marked in Fig. Ill, d was taken at 0-75inch diameter. KIn the eighth column from the left we get 7=, and this

V7Aremains constant throughout the rest of the working, theonly variable being 1c. Values of a can easily be worked onthe slide-rule, watching the decimal point in doing so.There may be two or three taper nose ribs between eachfull taper rib and a column for these can be inserted in thecorrect place, if necessary.In Fig. V, the nose of L.N.R. and the first taper rib havebeen laid out from the tables in Fig. Ill, from the L.E.to the front spar only, and the rib flange centre is shownrunning into the centre of the L.E. tube, the hatched portionshowing the fabric leaving the rib to pass over the L.E.tube. The top and bottom flanges of the rib can, of course,be run in tangential to the L.E. tube, and is a matter ofchoice of attachment of ribs to L.E. tube. In laying outthe wing tip in plan, the writer has shown radii in terms of

" C," the chord, trusting same will be useful to juniordraughtsmen when laying out a round wing tip. The formobtained will be found to give good results aerodynamically.All ordinates are above datum, and the datum in Fig. IIhas been purposely/ shown much lower than is found inpractice to show up a and y more clearly, but is truly repre-sented in Fig. L

Owing to lack of space, the summaries of AeronauticalResearch Committee Reports and Memoranda which usuallyappear in the last part of THE AJKORAIT ENGINBBB havehad to be held over this month. It is hoped to publish alengthy series of them in (he May issue.ED.

362A

The Aircraft Engineer April 24, 1931

Documents

Transcript of The Aircraft Engineer April 24, 1931