1911 aeriallocomotion00harprich

8/6/2019 1911 aeriallocomotion00harprich

1/184

IC-NRLF

AERIALLOCOMOTIONE.H, HARPER

A.NDALLANFERGL -

UNIVERSITY OF CALIFORNIA LIBRARY


2/184

LIBRARYOF THE

UNIVERSITY OF CALIFORNIA.Class


3/184


4/184


5/184

The Cambridge Manuals of Science andLiterature

AERIAL LOCOMOTION


6/184

CAMBRIDGE UNIVERSITY PRESS?Lon&Ott: FETTER LANE, E.G.C. F. CLAY, MANAGER

1*1

TOO, PRINCES STREETIonium: H. K. LEWIS, 136, GOWER STREET, W.C.

Berlin: A. ASHER AND CO.leipjtg: F. A. BROCKHAUS

ei3$orfc: G. P. PUTNAM'S SONSUombau anto (Calcutta: MACMILLAN AND CO., LTD.

All rights reserved


7/184


8/184

H

IaI


9/184

AERIALLOCOMOTION

E. H. HARPER, M.A.AND

ALLAN FERGUSON, B.ScWITH AN INTRODUCTION BY

G. H. BRYAN, Sc D., F.R.S

Cambridge :at the, University Press


10/184

Cambridge :PRINTED BY JOHN CLAY, M.A.AT THE UNIVERSITY PRESS.

With the exception of the coat of arms atthe foot, the design on the title page is areproduction of one used by the earliest knownCambridge printer, John Siberch, 1521


11/184

PKEFACETHIS little volume makes no claim to originality,its purpose being to present, in a manner assimple as is consistent with scientific accuracy, aconnected statement of the principles underlyingaerial locomotion ; and for this reason, whilst a certainamount of technical and scientific phraseology hasbeen employed, part of the introductory chapter hasbeen devoted to an elucidation of those mechanicalideas and terms, a correct comprehension of whichis absolutely necessary to the reader who desiresto follow intelligently the development of modernaeronautics. It will be noticed that some of thechapters 'overlap' to a slight extent a result whichit has not been thought worth while to avoid, asa second presentation of facts which may be some-what unfamiliar will not be without its value.

It becomes more and more difficult, as the numberof manuals and treatises on aeronautics continuallyincreases, to find a title which possesses the merit of223970


12/184

viii PREFACEfreshness. So far as we are aware, the title f AerialLocomotion ' has only previously be used in a paperby F. H. Wenham, published in 1866.The frontispiece to the book, and Figs. 13, 14, 20,and 21, have been supplied by the Topical PressAgency ; Fig. 22 is from an old print, and Fig. 26has been taken from Langley's Researches andExperiments in Aerial Navigation. Our thanks aretendered to Messrs Whittaker & Co. for permissionto reproduce several illustrations from the Englishtranslation of Moedebeck's Pocket-book of Aero-nautics.

E. H. H.A. F.

UNIVERSITY COLLEGE OF NORTH WALES,BANGOR.

April, 1911.


13/184

CONTENTSCHAP. PAGE

Introduction . . . . . xI. General Principles 1II. Propellers and Motors .... 28III. Stability and Control of Aeroplanes . 45IV. Model Aeroplanes and Gliders . . 66V. Aeroplanes . . . . . . 76VI. Dirigibles 94VII. Historical 119VIII. Conclusion . . . . . .151

Index 163


14/184

INTRODUCTIONBy Professor G. H. BRYAN, Sc.D., F.R.S.

THE accounts of aviation feats contained in news-papers do not afford the average reader muchinformation regarding the construction and workingof the machines on which these records have beenperformed. One day we may hear of an aviatorwinning a prize of 1000 for a flight from Land's Endto John o' Groats, another day we are told that threeItalian aviators have been killed. In some casesrecord flights are said to have been made on mono-planes, in others on biplanes, and the question is thusoften asked, which is better, the monoplane or thebiplane ? Frequently advantages are claimed for oneor other of these types, which do not really dependat all on whether the machine is a monoplane or abiplane, but which result merely from the fact thatsome particular machine of one or other type pos-sesses structural peculiarities which other machinesdo not possess. Such peculiarities may depend, forexample, on the position of the centre of gravity or ofthe propeller, and not on whether the machine hasone lifting surface or two superposed surfaces.

It is the object of Messrs Harper and Ferguson'sbook to give an intelligent general account of theconstruction and use of aeroplanes and of the way in


15/184

INTRODUCTION xiwhich they are adapted to the task they have toperform. A further useful purpose which the bookshould serve is that of showing the important partwhich has been played by indirect methods ofinvestigation in rendering aviation possible. Fromthe earliest days of history there has never been anylack of men who would only be too glad to buildflying machines and risk their lives by trying to flythem, but until quite recently the means availablefor accomplishing this object were inadequate. Thechange was brought about partly by a scientificstudy of the conditions of the problem, based in thefirst instance on determinations of the pressure of airon moving planes. This is a subject which hadattracted the attention of mathematicians long beforeaviation became possible, and two alternative theorieshad been proposed, one due to Newton and the otherto Helmholtz and KirchhofF. The next stage, theexperimental one, was mainly developed by the lateDr S. P. Langley who showed that the conditionsrequisite for aviation were more favourable thanhad been previously supposed. In this connection,however, the same old story repeated itself, whichhas so often retarded progress in discoveries andinventions, namely prejudice against the scientificexpert. The grant available for Dr Langley's ex-periments was withdrawn just at the time whensuccess was imminent. Had this not been the caseit is probable that the development of modernaviation might have proceeded along safer lines thanhas actually been the case. For it is certain thatthe models with which Dr Langley experimented


16/184

xii INTRODUCTIONpossessed better inherent stability than the majorityof existing aeroplanes, otherwise they would not haveperformed long flights without overturning in someway or other.The advantages of theoretical and indirect methodswere further illustrated in the case of aeroplanemotors. It was largely the development and im-provement of motors for use on roads which led tothe evolution of a motor sufficiently light andpowerful for the purposes of aviation. But it cannotbe denied that the construction of internal com-bustion engines has been greatly facilitated by thedevelopment of the science of thermodynamics whichlong preceded their introduction.

Although there is still a good deal of hostility onthe part of some practical men towards workers inpure science, there are nevertheless indications thata better state of feeling is beginning to arise, andthat those who are concerned with the constructionand use of aeroplanes are beginning to show greaterwillingness to pay attention to investigations of atheoretical character which may be likely to suggestdirections for further improvement. The presentbook will I hope show the general reader that theproblem of flight is a many-sided one and that it isonly by studying this problem from every possiblepoint of view that locomotion through the air can bebrought to the same final stage of development thathas been attained in the case of locomotion on landor water.


17/184

CHAPTER IGENERAL PRINCIPLES

THE conquest of the elements around him is,and ever has been, one of the objects for which manis continually striving. He early learned to supporthimself in water and to move rapidly over its surface,but the emulation of the birds, though dreamed ofsince the days of Dsedalus, has only of recent yearsbecome an accomplished fact. And now, when thesubject of artificial flight is so much, literally andmetaphorically, 'in the air,' when each day's news-paper has its account of some new record made, orold record broken, an elementary account of theprinciples underlying artificial flight, and the maintypes of machine in common use, may prove usefulto those readers who are not content to take factsas they stand, but wish to know something both ofthe ' why ' and the ' how ' of those facts.

It is however, impossible, to demonstrate clearlythese principles without some knowledge of thefundamentals of mechanics, and as an exposition

H. & F. 1


18/184

2 AERIAL L000MOTION [OH.of tfie points which chiefly concern us will, in thefuture, save much circumlocution, we give them hereby way of preamble.We shall constantly use the term 'force,' andsince in aeronautics, as in all scientific work, it is ofprimary importance to avoid looseness of terminology,we follow Newton's example, and define the term' force ' to mean ' that which alters, or tends to alter,a body's state of rest, or uniform motion in a straightline.'

If, then, any body under consideration be eitherin a state of rest, or moving with a constant speed, weinfer that no forces are acting on it. If, on the otherhand, its speed is varying either in magnitude or indirection, we infer that some force is acting on thebody and causing this variation. The importance ofthe phrase which we have just emphasised will bemade apparent in the sequel.

Now, in order to specify a force completely, werequire to know both its magnitude and its direction.A force equal to the weight of 10 Ibs. acting in anortherly direction is very different from a force of10 Ibs. weight acting in an easterly direction, for theone will urge the body on which it acts towards theNorth, the other towards the East. Bearing this inmind, we see that we can graphically depict anygiven force by means of a straight line, the length ofthe line, on some appropriate scale, representing the


19/184

I] GENERAL PRINCIPLESmagnitude of the force, and its direction the directionof the force. Thus in the diagram, the line AS maybe taken to represent a force of 10 Ibs. weight actingto the North, the line AC a force of 8 Ibs. weightacting to the North-East, and the line AD a force of

B

Fig. 1.

5 Ibs. weight acting to the East. Now, if only oneforce be acting on a body, that body will move withever increasing speed in the direction in which theforce is acting. But supposing that at some particularpoint of a body we have two forces applied ; how will12


20/184

AERIAL LOCOMOTION [CH.the body move under these circumstances? Theanswer is given by a very simple construction, whichthe reader is asked to bear carefully in mind, as weshall continually make use of it. Suppose A to bethe point in the body at which the two forces areapplied, and let the magnitude and direction of action

Fig. 2.

of one force be represented by the line AB, and ofthe other by the line AC.Draw the dotted line CD parallel to AB and BDparallel to AC, so that these two lines meet at D.Draw the straight line AD. Then the body willmove in the direction AD, and its motion will be


21/184

I] GENERAL PRINCIPLESsuch as if a single force of magnitude AD wereacting on it. (AD will be drawn on the same scaleas AB and AC.) And vice versa, any single forcesuch as AD acting on a body can be split up intotwo forces such as AS and A G, and the effect ofeach of these forces studied separately. In scientificterminology, AD is called the resultant of the forcesAB and A C, and the forces AB and AC are calledcomponents of the force AD.

C D

Fig. 3.

It is pretty clear that a force such as AD can bedivided into two ' components' in any number ofways, for any number of parallelograms can be drawnhaving AD as diagonal, but when the parallelogrambecomes a rectangle, as in Fig. 3, then A b and Acare called, par excellence, the components of theforce AD, and whenever we use the phrase l thecomponents' it will be with this signification.


22/184

6 AERIAL LOCOMOTION [OH.Now let us study in some little detail the motion

of a body which is acted on by several forces.Suppose, for example, we have a body acted on bytwo vertical forces P and Q, one acting upwards, theother downwards, the points at which these two forcesact being in a vertical line. Then the actual force

Fig. 4.

the resultant force on the body will be a force equalin magnitude to the difference between P and Q,and the body will move upwards or downwardsaccording as P is greater or less than Q. If P beexactly equal to Q, then, of course, the body will bein equilibrium.It is important to notice, also, that unless the


23/184

I] GENERAL PRINCIPLESpoints at which the two forces act are in a verticalstraight line, the conclusions which we have justreached will no longer be valid. If the two forces,which we will suppose to be equal in magnitude, actas shown in Fig. 5, then, so far from the body beingin equilibrium, it will turn round in the direction

Fig. 5.

indicated by the arrow, and the correct magnitude ofthis turning tendency is measured by the product ofone of the forces into a, the perpendicular distancebetween the two forces. In fact the two forcestogether constitute a turning system, which is knownas a l couple/At this juncture it is advisable to say a few


24/184

8 AERIAL LOCOMOTION [CH.words about the famous ' Third law of motion ' firstenunciated by Newton. It may briefly be expressedthus 'To every action there is an equal and oppositereaction.' For example, if we press our hands upona table, the table exerts an equal and opposite forceon our hands. If a horse in pulling a cart exerts adefinite force upon it, the cart exerts an equal andopposite force upon the horse. This equality ofaction and reaction has been made an occasion fora well worn and somewhat absurd puzzle-question' Why, if these two forces are equal, do the horse andcart ever get into motion at all?' The answer isobvious. In dealing with the motion of any body,we must ask ourselves what are the forces acting onthat body, and on that body only. The sum ofthese forces when added, or compounded, in themanner shown above will, in general, be a definiteforce acting in a definite direction, and the body willmove in that direction with ever-increasing velocity.Now, returning to our problem of the horse and cart,the forward pull of the horse is exerted on the cart,whilst the backward reaction of the cart is exertedon the horse, and has nothing to do with the motionof the cart itself. It is, indeed, a fact which is oftenlost sight of, that all forces consist of stresses betweentwo material bodies, and the Newtonian mode ofestimating forces tends to fix attention on one aspectonly of the stress.


25/184

i] GENERAL PRINCIPLES 9And now, leaving these somewhat tedious, but

very necessary, introductory principles, let us seehow they may be applied to the elucidation ofaeronautical problems.

Before we can propel ourselves through the air,we must first be able to rise into the air, and as iswell known this latter problem is solved by theuse of two types of machine the heavier-than-airmachine, and the lighter-than-air machine. The kiteis a representative of the former class, the balloon ofthe latter.

Let us deal with the latter type first, and seek ananswer to the question 'Why does a balloon rise fromthe ground ? '

Long before a balloon ever did rise 2000 yearsago the answer was supplied by Archimedes, whoshowed that when a solid is completely immersed inany fluid, it experiences an upward thrust which isequal to the weight of fluid displaced.

Think, then, of a solid substance immersed in afluid such as water. Two forces act on it its weightQ in a vertically downward direction (see Fig. 6), andthe upward force or thrust P, which by the principle ofArchimedes is equal to the weight of fluid displaced.Now if the solid, volume for volume, be heavierthan the fluid, Q will be greater than P ; the resultantforce will therefore be downwards, and the body will


26/184

10 AERIAL LOCOMOTION [CH.sink. If the solid be specifically lighter than thefluid, P will be greater than Q and the body will riseto the surface, while, if the solid and fluid have thesame specific gravities, the solid will rest indifferentlyin any part of the fluid.

And this gives us the reason why a balloon risesin the air. A cubic metre of air, under standardconditions of temperature and pressure, weighs T293kilogrammes. A cubic metre of hydrogen gas undersimilar conditions weighs only *088 kilogrammes; ifthen we take a light but gas-tight envelope of somekind and fill it with hydrogen gas, so as to form aballoon of say 1 c. metre capacity, the balloon will


27/184

i] GENERAL PRINCIPLES 11rise in the air when set free, since there will be aresultant upward force on it equal to the weight

of1*205 kilogrammes : the balloon could therefore liftthis weight from the ground, and whatever be theweight of the whole system, envelope, hydrogen,supports, and car, so long as it does not exceed1*205 kilogrammes, the balloon will rise until itreaches a point where the atmosphere is of such atenuity that the weight of the displaced air is justequal to that of the whole balloon system ; it willthen remain in equilibrium.And now, having raised ourselves into the air, wehave next to ask ourselves how we may direct ourballoon through the subtle fluid in which it isimmersed.

This problem has two aspects the propelling ofthe balloon, and its steering. But it is imperative tonotice that these two problems are intimately con-nected, in so far that it is utterly impossible to steerthe balloon unless we have some means of propellingit of giving it an independent speed through the air.If you are afloat in a rowing-boat and lose the oars,then, work the rudder as you will, the boat remainsthe sport of every chance current. It is only whenthe oars are being used that the rudder has anyeffect. And so it is with our balloon. Let us thenturn our attention to the question of propulsion, andas a simple preliminary illustration, let us consider


28/184

12 AERIAL LOCOMOTION [CH.the dynamical principles involved in that verycommon and every-day process called walking. Inwalking, one presses one's foot hard against theground ; the earth exerts an equal and oppositepressure on the foot, and this force, which is exertedon the walker, propels him forward. And in just thesame way the screw propeller of an airship, whirlinground rapidly, beats the air backwards, exerting abackward force upon it. The reaction of the airupon the propeller is, by Newton's third law as wehave explained, a forward force which is exertedupon the propeller, and this forward force is trans-mitted to the body of the airship and urges itforwards.

And now, having seen the principles which governthe raising of a balloon in the air, and its motionthrough the air, we have to turn our attention to aconsideration of the leading principles which governthe motion and flight of aeroplanes the heavier-than-air machines.

The general laws on which the aeroplane dependswere explained just over one hundred years ago bySir George Cayley. In Jules Verne's and similarstories wonderful airships are described as held upin the air by screws turning about vertical axes.One of the difficulties in realizing this dream is theweight of the engines that would be required to worksuch propellers. Until an engine has been constructed


29/184

I] GENERAL PRINCIPLES 13giving much more power for its weight than thelightest yet made, such an airship could not risefrom the ground. Sir George Cayley showed howby using the screw to drive the airship forward itcould raise and support in the air a much greaterweight than if it were used directly to lift it, aseemingly paradoxical result.

Fig. 7. AE direction of motion of inclined plane,BA air-pressure, CA Lift, DA Drift.Suppose a flat surface is driven through the air

horizontally, the front of the surface being raised alittle higher than the rear, so that the under side strikesthe air as it advances (see Fig. 7). The air will exert apressureon the under side at right angles tothesurface.This pressure may be considered to consist of two parts


30/184

14 AERIAL LOCOMOTION [CH.or components, one, vertical, pushing the planeupward called the Lift, the other, horizontal, pushingit back called the Drift. If the surface is nearlyhorizontal, the front being raised only a little abovethe back, the air pressure will act almost verticallyupward ; so the Lift will be much greater than theDrift. The Lift acting upward supports the weightcarried by the plane, while to keep the plane movingthe Drift must be counterbalanced by an equal forcedriving it forward. This force is obtained by theaction of the screw propeller. The weight carried,being equal to the Lift, will be many times greaterthan the propeller thrust which is equal to the Drift,while in a direct-lift machine or kelicoptere thepropeller thrust would have to be the same in amountas the weight. Even so the lightest engines inSir George Cayley's time and for nearly a hundredyears after were too heavy to be successful. Theproblem before inventors then was to design anaeroplane giving the greatest possible Lift in pro-portion to Drift and to make it both strong and lightso as to be able to carry a heavy engine. The engineitself had to be made as powerful as possible inproportion to its weight. Only by the development,in the first instance in connection with the motor-carindustry, of the light-petrol motor has flight byheavier-than-air machines been rendered possible.All bodies moving through the air have a pressure


31/184

I] GENERAL PRINCIPLES 15exerted on them which, as before, may be resolvedinto Drift or force resisting motion, and Lift or forcepushing the body at right angles to its direction ofmotion. When there is very little Lift or none at allthe Drift, which is then practically the entire airpressure, is called also Head-resistance. Since theweight of an aeroplane is entirely supported by theair, and the engine and propeller of both airships andaeroplanes are necessary to counteract the Drift orHead-resistance, we have next to consider the funda-mental principles of air-pressure on bodies movingthrough the air.

First of all the air-pressure depends only on therate and direction with which the air and the bodymeet ; it is the same whether the body moves tomeet the air or remains still while the air blowsagainst it. This fact is sometimes expressed by sayingthat a body moving through the air creates an artificialwind. If the air is calm and an aeroplane flies at40 miles an hour eastward, the air-pressure is the sameas if the wind blew at 40 miles an hour from the east,the aeroplane facing it and remaining at rest ; or as ifthe wind were 10 miles an hour from the east whilethe aeroplane moved eastward at 30 miles an hour ;or again as if a west wind of 10 miles an hour wereblowing and the aeroplane moved eastward at 50 milesan hour. In all these cases the aeroplane meets theair at a rate of 40 miles an hour.


32/184

16 AERIAL LOCOMOTION [OH.Obviously the greater the velocity with which the

aeroplane and air meet, the greater will be the air-pressure. The weight carried by the aeroplane isequal to the Lift. If the speed through the airincreases, the Lift will increase and, becoming greaterthan the weight, will push the aeroplane upward. Ifthe speed diminishes, the Lift will not be sufficientto support the weight and the aeroplane will beginto come down. For these reasons an aeroplane hasa certain speed at which it flies through the air andit cannot be made to fly at different rates, at leastnot without altering its shape or the position of someparts which at present can only be done to a verysmall extent. This speed is speed through the air,for on that the air-pressure depends. If an aeroplane'sair-speed is 40 miles an hour, then flying with a windof 30 miles an hour it would have a speed of 70 milesan hour over the ground, and flying against the samewind a speed of only 10 miles an hour over theground.

How the air-pressure or Drift and Lift on anythingvaries with the shape of the body, its speed throughthe air, and the direction in which the air meets itis a question which can only be studied experimentally.Bodies of different shapes and sizes are moved orheld in different positions and the velocity with whichthe air meets them,the Lift and the Drift, are measured.Precautions must be taken to ensure that the velocity


33/184

il GENERAL PRINCIPLES 17and the direction of the motion with which the airand body meet are really known and that thesequantities do not change while the experiment isbeing made. On this account rather elaborate andexpensive apparatus is required to make the measure-ments with convenience and accuracy. ProfessorLangley was enabled by a grant from the UnitedStates Government to carryout a series of experiments,and as full records of these were published they havebeen ofthe greatest service to all interested in aviation.In England, Sir Hiram Maxim was in the fortunateposition of being able to devote a large sum of moneyto experimental work, and has published some of hisresults. The British Government has lately taken upthis work and experimental researches are carriedout at the National Physical Laboratory.

Three principal methods have been used. Phillips,Langley, Maxim and others employed a long beam,usually called a whirling table, to carry the bodyround in a horizontal circle. The body may also beheld in the centre of a large pipe through which afan drives a current of air. In this case precautionsmust be taken to ensure that the air flows steadilyand uniformly through the pipe and the body mustoccupy only a small space in the centre. Anapparatus of this sort is in use at the NationalPhysical Laboratory. Experiments have been madealso on bodies sliding down wires or along rails, a

H. & F. 2


34/184

18 AERIAL LOCOMOTION [OH.method which seems to be particularly useful whenonly Drift or Head-resistance is to be measured.Rougher experiments have been made with bodiesattached to engines or motor-cars, or flown as kitesor held exposed to the wind.

Experiments in which Lift and Drift or Head-resistance are measured are of primary importance ;but to explain the different effects on bodies ofdifferent shapes we must adopt other classes ofexperiments. In these the disturbance produced inthe air is studied. The contribution of each part ofthe surface to the air-pressure has been measuredseparately by connecting a small hole in the surfaceby a tube to a manometer. Bits of thread attachedto the surface show how the air passes over it, andsome information can be obtained by holding theflame of a candle in the air. Again, if the air is madevisible by being mixed with smoke or steam it can bephotographed.Some general principles have now been established.To diminish the Head-resistance, all bodies movingthrough the air should be as smooth as possible,without projections or sharp edges, shaped so as topart the air gradually in front and allow it to closein again gradually behind. It is found that anythingin the way of a sharp edge, a projection, or an abruptlyending stern, causes eddies behind it and greatly in-creases the Head-resistance (Fig. 8, B). It is found


35/184

I] GENERAL PRINCIPLES 19best to have a stern gradually tapering to a point forat least two-thirds of the whole length. The bowmay be blunter and the widest part should not bemore than one-third of the length from the front.' The head of a cod and the tail of a mackerel ' is now

BFig. 8. A. Air-current passing over body of approximate stream-

line form. B. Eddies produced by abrupt stern.

a commonplace expression in books on aviation.This shape is often called the stream-line form, becausethe air parting in front of the body streams over it andcloses up behind without any eddying or any draggingof the air with the body (Fig. 8, A). The bodies offish and of birds are shaped on this plan and closely22


36/184

20 AERIAL LOCOMOTION [CH.resemble each other. Birds are however broader inproportion to their length than fish, and this is dueto the difference in the streamline form for water andfor air. This principle is followed in the constructionof the gas-bags of dirigible balloons, but the con-structors of aeroplanes seem to think that the labourof shaping struts etc. in this way would not berecompensed by the advantage gained in loss ofHead-resistance.Two objects close together experience a muchgreater Head-resistance than when far apart, and whena number are close together the air in passing throughthe spaces between them causes a very great Head-resistance. In this case it is better to enclose theobjects in a covering which presents a single smoothsurface to the air, the covering of course beingshaped as far as possible according to the principlesenunciated in the preceding paragraph. In someaeroplanes the aviator, engine etc. are enclosed ina boat-like structure of this sort.

The best form for the supporting surfaces ofaeroplanes is a very complicated question. First ofall we have the principle of Aspect Ratio. Whenthe air produces Lift on a part of the surface it ispressed down by that part of the surface at the sametime, and so must have a downward motion communi-cated to it. This air, since it is now moving in aslightly downward direction, cannot be so useful in


37/184

i] GENERAL PRINCIPLES 21producing a Lift or upward pressure on that portionof the surface with which it comes into contact next.Receiving a further downward motion it will be stillless efficacious on the part of the surface behind that,and so on. The front portions of the plane, therefore,produce most Lift, and at a certain distance from thefront a continuation of the surface produces hardlyany Lift at all. For this reason supporting planesnarrow from front to rear with a long front edge givethe best result. The ratio of the breadth from frontto rear to the length parallel to the front edge is calledthe Aspect Ratio. In present-day aeroplanes theAspect Ratio varies from 1 in 5 to 1 in 8.To counteract the loss of Lift due to the downwardmotion of the air the back of the plane may be bentdown so that the air still presses on it. There is alimit to the extent to which this may be done. Theair-pressure now is nearer the horizontal, and theproportion of Drift to Lift is also increased. At acertain point the disadvantage of increase of Driftcounterbalances the advantage of increase of Lift ; sowe are led again to the principle of Aspect Ratio.The air also has a tendency to escape sidewise at theends of the plane, and this loss will be greater thegreater the breadth of the plane.

Surfaces curved from front to rear with theconcavity downward are found to give a largerLift in proportion to Drift than flat surfaces, and


38/184

22 AERIAL LOCOMOTION [CH.consequently the supporting surfaces of aeroplanesare of this form. Phillips was the first to advocateconcave surfaces and in 1884 patented some sectionswhich he had found most advantageous. Lilien-thal, who is mentioned later in connection withgliders, knew the advantages of concave wings.Different explanations of their efficiency are given.

BFig. 9. A. Gradual deflection of air-current.B. Air-current meeting dipping front edge.

When the air strikes the front edge of a flat surfaceon the under side, it must receive a somewhat suddenshock, the resulting downward motion accounting forthe Lift as previously explained. It is generally heldthat a better effect is obtained by communicating thedownward motion to the air gradually instead ofwith a shock. This principle is assumed to hold also


39/184

i] GENERAL PRINCIPLES 23in the similar cases of windmills, rotatory fans, screwpropellers and turbines. According to this explanation,at the front edge the air should pass along the surfacetangentially without striking it and be graduallycast down by the curving surface behind (Fig. 9, A).The curving of the rear part of the surface has beenmentioned before in connection with Aspect Ratio.Phillips on the other hand advocates the 'dipping frontedge ' (Fig. 9, B). He says the curvature should be sogreat at the front that the air strikes the upper sideof the surface and is diverted upwards ; this upwardmotion, according to his theory, leads to diminishedpressure on the upper side and consequently to Lift.The form of the upper surface certainly seems tohave something to do with the amount of Lift, for asection with the under surface flat and the upperconvex gives a good ratio of Lift to Drift. This andmany other results cannot be considered as fullyexplained by either of the theories given above, andfurther study of the motion given to the air in thesecases is evidently required. It seems to be settledby experiment that almost flat surfaces with verysmall curvatures, give the best ratio of Lift to Drift,while surfaces with a larger curvature give a largerLift in proportion to the size of the surface, but notso good a ratio of Lift to Drift. Some aeroplaneconstructors prefer the latter form on account of thesmaller surface required for support, though more


40/184

24 AERIAL LOCOMOTION [OH.engine-power is required to counterbalance thegreater Drift.

In a description of aeroplane wings or supportingsurfaces the edge of the surface which first meets theair is called the front or entering edge. The edge atwhich the air leaves the surface is the trailing edge.A straight line from the front edge to the trailingedge is called the chord. The length of the surfacein the direction of the front or trailing edge is the

Fig. 10. ADB section of curved surface, BF direction of motion,A entering edge, B trailing edge.span. The Aspect Ratio, as explained before, is theratio of the chord to the span. The distance of apoint on the surface from the chord is the camberand the position of maximum camber is where thedistance of the surface from the chord is greatest.The angle of incidence is the angle which the chordmakes with the direction in which the plane movesto meet the air ; the angle of entry the angle whichthe tangent to the section of the surface at the


41/184

i] GENERAL PRINCIPLES 25entering edge makes with the chord ; and the angleof trail is the corresponding angle at the trailingedge. In describing the curvature of the surfacefrom front to rear, the angles of entry and trail, theamount and the position of the maximum camberare generally given.The wings must be light, and yet strong enough tobear the air-pressure without being deformed. Theyare made of fabric of some sort stretched on andsupported by a framework built up of two main sparsin the direction of the span crossed by ribs at frequentintervals. The framework may be covered on oneside only, when the wing is termed single-surfaced.To diminish Head-resistance, the spars and ribs in thiscase are enclosed in pockets of fabric sloping downto the single surface. If the framework is coveredon both sides, the wing is termed double-surfaced.In this case the distance between the upper andlower surfaces may be large consistently with goodresults, and so the main spars may be built-up girdersof some depth. These give greater stiffness to thewing and so diminish the number of supporting staysand wires which would otherwise be necessary.Wenham, as the result of observations on birdsflying one above another, propounded the principlethat two supporting surfaces, such as are used inaeroplanes, do not materially interfere with eachother's action if the vertical distance between them


42/184

26 AERIAL LOCOMOTION [CH.is not less than the breadth of either. Instead of usingone surface it may be broken up into two or moreplaced vertically over each other, a great advantagewhen the surface is long. In addition, by connectingthe main spars of two such surfaces by vertical strutsand diagonal tie-wires "they can be made to supporteach other and the resulting structure is very stiffand strong. An aeroplane constructed in this way iscalled a biplane. If there is only one supportingsurface the aeroplane is a monoplane, if three a tri-plane. Phillips constructed an apparatus in whichthe principles of Aspect Ratio and Superposed Planeswere carried to excess. He used 50 planes each 22feet by 1^ inches placed 2 inches apart and showedthat a considerable Lift could be obtained.The shape of the wings of birds may be adducedin support of these principles. The outspread wingsof birds have a large span in proportion to theirbreadth, are concave beneath, and have the steepestpart of their curve close to the entering edge of thewing. That the body of a bird is so shaped as toexperience least resistance from the air has beenalready mentioned. Birds that soar, that is, fly withoutstretched motionless wings and seemingly withouteffort, illustrate these principles best, for their flightpresents the greatest resemblance to that of anaeroplane.A kite is supported by the air in the same way as


43/184

i] GENERAL PRINCIPLES 27an aeroplane. The kite is held at rest and the airblows against it, but the effect produced is exactlythe same as if the kite moved to meet the air andthe air remained at rest. Lift and Drift, then, areproduced by the air and in this case are counter-balanced by the pull of the string on the kite. Thelarger the Lift in proportion to the Drift, the nearerto the vertical will be the string ; while the smallerthe Lift in proportion to the Drift the more nearlyhorizontal will be the string where it is fastened tothe kite. It is easier to construct a kite that will flythan an aeroplane, because the ratio of Drift to Liftdoes not require to be so small, but the Lift ofcourse must be large enough to raise the weight ofthe kite and some string as well. The string of akite may be regarded as taking the place of theengine and propeller of an aeroplane as regardsweight and propeller thrust, but without the limitationto which the propeller thrust to be obtained from anengine of given weight is subject. That kite whichhas the greatest Lift in proportion to Drift willrise highest and nearest the vertical, and so theconstruction of a good kite and of an aeroplane arevery similar problems.


44/184

CHAPTER IIPROPELLERS AND MOTORS

1. PropellersTHE principle of action and reaction states that a

force, tending to move a body in one direction, musthave corresponding to it a force tending to movesome other body in the opposite direction. An air-ship or aeroplane is completely surrounded by air ;in consequence the only other body on which thereaction, corresponding to the force propelling theairship or aeroplane forward, can act is the air itself.The aeroplane is driven forward and the air backwardat the same time. To drive the air back a screwpropeller is used.Any screw, that of an office press or of a liftingjack for example, consists of a central axis with aprojection called the thread running spirally roundit. The screw fits into a nut in which a groove hasbeen cut corresponding to the thread of the screw.Turning the screw round causes the thread to pressagainst this groove and produces a force tending to


45/184


46/184

30 AERIAL LOCOMOTION [OH.The air however cannot remain still, for the pressureof the propeller blades casts it back. As a result theair is sucked in from the front and all round. Evidentlythe propeller cannot cast back the air any faster thanit could a solid nut. So if we suppose the air afterbeing cast back to move as a fixed nut would, wetake the most favourable view possible of the actionof the propeller. This would make the pitch of thescrew equal to the distance the screw moves forwardadded to the distance the air moves back during onerevolution. Stated in another way, the forwardvelocityof the propeller added to the backward velocity com-municated to the air is equal to the velocity withwhich the propeller would move turning at the samerate in a fixed nut. The velocity communicated to theair is called the slip of the screw. It is clear thereforethat the pitch of a screw propeller must be greaterthan a certain amount to be of any use at all. Forexample, an aeroplane moves at the rate of 45 milesan hour, that is, 3960 feet in a minute. If the pro-peller makes 1000 revolutions a minute, the aeroplanemoves forward nearly four feet while the propellermakes one revolution, and consequently the pitch ofthe screw must be more than four feet. At the sametime the pitch must not be too great ; if it were, theblades would stir the air round instead of casting itback. For each particular speed of the airship andspeed of revolution of the propeller there will be one


47/184

ii] PROPELLERS AND MOTORS 31pitch which will give the best result, and this canonly be determined experimentally.

The revolving blades trace out a circle whosediameter is called the diameter of the propeller, andthe air they reach is set in motion. Again, takingthe most favourable supposition, the air cast back isa cylindrical column whose cross-section is the circlereferred to. Now the principles of mechanics showthat the force, driving the aeroplane forward and theair back, is proportional to the quantity of air castback in any time multiplied by the velocity givento it. The work done in casting the air back, andconsequently the amount of the power of the engineexpended in moving the air, is proportional to thesame quantity of air multiplied by the square of thevelocity. Now this part of the power of the engine,usually said to be expended in slip, is wasted as faras driving the aeroplane is concerned, and should bemade as small as possible. It varies as the square ofthe slip, so the slip should be made as small as possible.To keep up the propelling force the quantity of aircast back must be made large, and so the diameterof the propeller is large. Propellers of large diameterand small slip are, therefore, most economical of power.If the diameter of the propeller is made smaller, theslip must be increased to produce the same thrust.As already pointed out, this cannot be done withadvantage by increasing the pitch of the screw, leaving


48/184

32 AERIAL LOCOMOTION [CH.its rate of revolution unaltered. The speed ofrevolution must be increased. To produce the samethrust propellers of large diameter and slow speedof revolution may be used, or propellers of smalldiameter revolving more rapidly. The former havethe smaller slip and are more economical of power.

This is, however, not the only consideration to betaken into account. Large propellers are heavier,and the weight puts a limit on the size of propellerthat can be used. Again, if the speed of revolutionof the propeller differs from that of the motor, thetwo must be linked up by some system of gearing bywhich the power of the motor is transmitted. Inthis transmission a certain amount of power is lost.For each size of propeller the power lost in slip, thepower lost in gearing, and the disadvantage of weightand size must be estimated, and that propelleremployed which is on the whole most advantageous.The petrol motor used on airships and aeroplanes hasa very high speed of revolution, over 1000 revolutionsper minute as a rule. If the propeller is mounteddirectly on the crank-shaft of the motor, it has asmall diameter ; thus a considerable amount ofpower is lost in slip but none in gearing. Suchpropellers are employed on almost all aeroplanesand on some airships. A few aeroplanes and someairships have larger propellers connected to the motorby gearing.


49/184

n] PROPELLERS AND MOTORS 33The blade of a propeller is not usually a simple

screw of the same pitch throughout. A better effectmay be obtained by varying the pitch as the distancefrom the axis increases, and also varying it acrossthe blade from front edge to back. Experiment ishere the only guide. By experiment, too, the bestbreadth of the blade must be found as well as thenumber of blades to be used. Attention may becalled to some important points connected with thetesting of screws. The explanation given aboveshows how much the motion of the airship throughthe air has to do with the eifect of the propeller.While being tested, the propeller should move throughthe air at the speed of the airship. As in testingsupporting planes etc., two methods are available :the propeller may be mounted, at the end of a rotatingarm for example, so as to move through the air ; orthe propeller, remaining still, may act in a current ofair blowing at the same speed in the opposite direction.Many tests on model propellers have been made. Alarge rotating arm has recently been erected at Barrowon which propellers of the largest size may be tested.Makers of propellers in advertising their wareshave often published statements of the thrusts ob-tained at a certain number of revolutions per minute.These statements were quite valueless for two reasons.In the first place the tests were made with propellersat rest in still air. In the next place, the thrust at

H. & F. 3


50/184

34 AERIAL LOCOMOTION [CH.a certain speed of revolution is a matter of no im-portance ; a very large thrust, if obtained by usinga very powerful and heavy motor, is no advantage.The thrust per horse-power of the motor used is theimportant thing, according to which the merit of thepropeller ought to be estimated.

It has been said that the best form of blade ina propeller is a matter for experiment. Some generalprinciples must be adhered to, very similar to thosewhich govern the form of the supporting surfaces ofaeroplanes. The blade of the propeller is to movethrough the air casting it back, instead of down ;and the air-pressure on it can be resolved into twocomponents, one parallel to the axis and one per-pendicular to it. The former is the useful component,pushing the propeller forward ; the latter is notuseful, but resists the motion of the blade and hasto be counteracted by the action of the motor, andtherefore its ratio to the useful component has tobe made as small as possible. This is almost thesame as the problem of Lift and Drift, and the samegeneral principles apply. The fact that the propellerblade moves faster at the tip than nearer the axis,alone distinguishes the propeller blade from thesupporting surface. First, suppose the blade shapedof such a pitch that it will move through the airwithout disturbing it, as if the air were a fixed nut.In this case, like a horizontal surface moving edge-


51/184

n] PROPELLERS AND MOTORS 35wise through the air, it meets the air edgewise andexperiences no thrust. Now alter the shape of theblade so that the air strikes it on the back and isdeflected backwards. In accordance with the principleof Aspect Ratio the blades should not be too broad ;in accordance with that of Superposed Planes thereshould not be too many blades, otherwise they inter-fere with one another's action. The blades should beconcave at the back and convex at the front withboth surfaces smooth. The principle of the dippingfront edge or the principle of the gradual deflectionof the air without shock will apply equally to thepropeller blade and to the supporting surface of theaeroplane.

Propellers are made of both wood and metal.Wooden propellers are built up of strips of woodrunning from the axis to the tip of the blade. Thestrips are first fastened together and afterwardsfashioned to the desired shape. In metal propellersthe blade is formed of a sheet of metal supportedby a rib running out from the axis.In ships the propeller is always placed behind theship, never in front. There are two reasons for this.First, if the propeller were in front, the water thrownback would strike the ship and impede its progress.Secondly, on account of the friction against the sides,the water at the stern is dragged forward to someextent, with the ship ; the propeller if placed behind32


52/184

36 AERIAL LOCOMOTION [CH.has first of all to destroy the forward motion of thewater, before it can give it a backward velocity.This velocity is then smaller and the work expendedin slip is therefore less, and consequently the propelleracts more efficiently. The same considerations applyin the air, but are of less importance and are generallyoverridden for constructional reasons in airships andaeroplanes. The propellers are usually placed behindthe main planes in biplanes, but in monoplanes onaccount of the presence of the boat-shaped bodywhich carries the tail the more convenient positionis in front.

2. MotorsThe importance of a light and efficient motor

in aviation has been emphasized elsewhere. Fora long time the steam engine was the only availablemotor, and many attempts were made to constructa light steam engine. Langley and Maxim eachachieved a certain degree of success, the former on asmall scale, the latter on a large one. In both cases,however, supplies of fuel and water for short flightsonly could be carried. The difficulties which havemade complete success impossible arise from thevery nature of the steam engine. The steam has atemperature very much lower than that produced bythe combustion of the fuel, therefore a large part of


53/184

n] PROPELLERS AND MOTORS 37the energy of the fuel is necessarily wasted in passingto the steam and a large supply has to be carried.Further, if a non-condensing engine is employed, as isdone on railways, a large supply of water is requiredfor a long flight, while if a condensing engine isemployed, so that the same water can be used overand over again, the weight of the condenser and thewater in it is considerable. The boiler also adds tothe weight. The development of the internal com-bustion motor in connection with the motor-carindustry opened up new possibilities. For in it noboiler or water for conversion into steam is required.The gases produced by the combustion of the fuelare used to drive the piston. These gases are atso high a temperature that they would destroy anypipes or valves they passed through ; the combustiontherefore takes place in the cylinder itself. Thefuel is converted into gas, mixed with air in thecorrect proportions for complete combustion whichtakes place instantaneously, i.e. as an explosion, whenthe mixture is fired by an electric spark. This directuse of the products of combustion is the chief causeof the superiority of the internal combustion motorover the steam engine. The gas is cooled by its ownexpansion as it drives back the piston and is ableto escape from the cylinder without destroying thevalve.The internal combustion motor is usually a four


54/184

38 AERIAL LOCOMOTION [OH.cycle one, that is, work is done by the piston onlyduring one stroke in every four. Suppose at thebeginning of a down-stroke an explosion takes place,the piston is driven down and work is done. Duringthe next up-stroke, the exhaust stroke, an exhaustvalve is opened through which the piston drives thegas that has done the work. This valve is thenclosed. A down-stroke, termed the admission stroke,follows, during which an admission valve is openedand the mixture of vaporised fuel and air suckedinto the cylinder. Then all the valves are closed andin the next up-stroke (the compression stroke) themixture admitted is compressed at the upper end ofthe cylinder. Now the mixture is ignited by anelectric spark, and another useful stroke follows. Tokeep the engine working steadily, if it has only onecylinder, a heavy flywheel is used. This acts as areservoir of power, storing up energy during theuseful stroke and giving it out again to keep theengine running during the three idle strokes. Theuse of a large number of cylinders enables the fly-wheel to be dispensed with or at least reduced insize. The cylinders ought to be arranged so thattheir useful strokes begin successively at equalintervals. The different moving parts should bebalanced to avoid vibration, the motion of one parthi one direction being counterbalanced by the motionof other parts in an opposite direction. For example,


55/184

n] PROPELLERS AND MOTORS 39with two cylinders their pistons should always bemoving in opposite directions. Placing the two onopposite sides of the crank shaft, the advantages ofbalance and of working strokes at equal intervals canbe combined, every alternate stroke being a usefulone on the part of one of the cylinders, but half thestrokes would still be idle ones for both cylinders.At least four cylinders must be used to make theiruseful strokes continuous, one beginning when anotherstops, and eight cylinders are common in aeroplanemotors.

The fuel used is petrol which contains the lighterconstituents of mineral oil and vaporises readily atordinary temperatures. Sometimes the vaporisingand mixing are done in the cylinder, the petrol beingsprayed in by a pump. The mixing however is notwell done by this method. More usually theseoperations take place in a separate vessel calleda carburetter. This must work regularly, alwayssupplying to the cylinders air and petrol vapourmixed thoroughly and in the correct proportions.The valves must be reliable. They must bestrong enough to remain closed during the explosion.The admission valve must admit a sufficient quantityof the mixture, and the exhaust valve must allow thespent gases to escape readily without unduly impedingthe motion of the piston. These valves only openduring the alternate down- and up-strokes respec-


56/184

40 AERIAL LOCOMOTION [OH.lively ; accordingly a separate shaft, called the camshaft, has to be provided to operate them, gearedso as to revolve once for two revolutions of thecrank-shaft.

The construction of a reliable apparatus forproducing the electric spark to ignite the mixture isa separate problem.The cylinder, piston, and valves have to be keptcool, for the expansion of the gases cannot be carriedso far as to prevent these parts from being undulyheated. The cooling is effected by passing over thema current of either air or water. In aeroplanes thepassage of air over the cylinders as the aeroplanemoves through the air is sometimes relied on. Thismethod seems to be efficient for a short period only,unless, as in the Gnome engine, to be referred tolater, the cylinders themselves revolve. Water coolingis more general. A jacket encloses the cylinder andwater circulates through the narrow space betweenthe jacket and the cylinder wall. This water passes toa radiator where it gives out its heat to the air andreturns cool to the cylinder to go round the samecircuit again.The parts of a petrol motor move very rapidly ;the crank-shaft usually makes more than 1000revolutions per minute. This rapid motion wouldsoon heat and destroy two parts bearing on eachother unless they are well lubricated so as to banish


57/184

n] PROPELLERS AND MOTORS 41friction. A system by which lubricating oil is auto-matically conveyed to all bearing surfaces is thereforenecessary.

Motor-car engines have by years of successiveimprovements been made very reliable. For airshipsengines of the same sort are generally employed,extreme lightness not being so important as reliability.Such engines however are too heavy for aeroplanes,so special engines, lighter in proportion to theirpower, have to be constructed. To reduce the weight,materials very strong and light may be used regard-less of their cost or of the expense of working them.All parts not absolutely necessary may be dispensedwith and the weights of all other parts cut down untilvery little margin of safety is left. Many aeroplaneengines are only motor-car engines specially lightenedin this way a method which gains lightness at theexpense of reliability. A better way to attempt thesolution of the problem is to construct an engine of adifferent pattern, specially suited to aeroplane work.A radial arrangement of the cylinders, i.e. an arrange-ment round the crank-shaft like that of the spokes ofa wheel round the hub, seems the most suitable. Allthe pistons here may be connected to the same crankan arrangement which saves a great deal of weight.If the number of cylinders is odd and the explosionsoccur in succession in alternate cylinders, first, thenthird, then fifth, and so on, the intervals between


58/184

42 AERIAL LOCOMOTION [OH.successive explosions will be equal. In many enginesa partial radial arrangement of the cylinders has beenadopted but cooling has been the chief difficulty toovercome. Sometimes the cylinders revolve while thecrank-shaft remains fixed, the action of the pistonsremaining the same. The rotation of the cylindersthrough the air keeps them cool. Centrifugal forceof course acts on all rotating parts, and with rotatingcylinders different arrangements have to be made foradmission and exhaustion and for lubrication. Themixture of petrol vapour and air has sometimes beensupplied through the hollow crank-shaft and pistonrod, the admission valve being placed in the piston.Motors of the rotary type have the disadvantage thata considerable amount of power is wasted in whirlingthe cylinders through the air. However, one of them,the Gnome, is generally regarded as the most power-ful and reliable aeroplane engine of the present day.Though it does not now hold the record for length offlight, more long flights have been made with it thanwith any other engine. Lubrication seems to be adifficulty, and after a day's flying it is said to requiretaking to pieces completely for cleaning.

Suggestions have been made for a non-revolvingradial engine, cooled by being placed in the currentof air produced by the propeller.

Anyone reading accounts of flights and of aviationmeetings in the press soon sees that aeroplane motors


59/184

ii] PROPELLERS AND MOTORS 43are very unreliable. The ignition, the carburetter,the lubrication, the cooling, and the valves all givetrouble. The failure of any of these, if it does notstop the engine altogether, causes its power to fall offand very probably brings the flight to an end. In factthe perfect working of a motor for more than a shorttime is sometimes mentioned as if it were a piece ofextraordinary good luck. Reliability trials are veryurgently needed. Mr P. Y. Alexander offered aprize for a British engine which could run unattendedfor 24 hours and fulfil certain other conditions. Theresults of the tests in this competition have recentlybeen announced. Only one engine, the Green, ranfor the twenty-four hours but it did not yield the fullpower required. Of other British engines long flightshave been made by the E. KV. motor towards theclose of 1910. The reliability runs which aided somuch in the development of motor-car engines maybe recalled. In these all breakdowns of the enginewere counted against it. Similar trials in which anaeroplane makes flights on successive days, therepairing and overhauling of the motor being eitherforbidden or carried out under official observation,ought to lead to great improvements in aeroplanemotors. Such trials are needed before the aeroplanecan become a vehicle for either popular or militaryuse. At present an aviator has to employ a staff ofmechanics and keep a reserve of engines and spare


60/184

44 AERIAL LOCOMOTION [CH. nparts, and repairs and overhauling are carried outafter almost every flight.

If the operations of exhaustion and admission canbe carried out effectively during a single stroke ofthe piston, the number of working strokes can bedoubled. The engine becomes a two-cycle one, andeach cylinder in it does as much work as two in afour-cycle engine. Other more visionary develop-ments depend on the discovery of some metal oralloy which would not be destroyed by the very hotgases resulting from the combustion of the fuel. Ifsuch materials can be found, the combustion cantake place outside the cylinder, the gases passinginto the cylinder as steam does in a steam engine.A motor turbine even could be constructed to beactuated by these gases. An electric accumulator,light enough to be a practicable source of power foraeroplanes, is not to be expected in the immediatefuture.


61/184

CHAPTER IIISTABILITY AND CONTROL OF AEROPLANES

1. Stability

ALTHOUGH an aeroplane is balanced correctly,it may fly only a short distance. After some time itmay overturn in either of two ways tilting head upor down, or canting sideways. The upsetting maybe completed in a continuous swerve of the aeroplanefrom its correct position in flight, or may be precededby pitching or rolling motions which become moreand more violent. This tendency of the aeroplaneis not due to any fault of balance and cannot becured by any adjustment of the weight. An aero-plane which flies steadily, without showing thistendency to upset, is said to possess stability. Ithas longitudinal stability when it has no tendencyto upset by tilting head up or down, and lateralstability when it has no tendency to upset by cantingsideways.There are three ways of obtaining stability. The


62/184

46 AERIAL LOCOMOTION [CH.aeroplane may be so designed that it returns of itselfto its correct position after any departure from it.This is called inherent stability. Or the pilot maybring the aeroplane back to its correct position ; inthis case stability is obtained by the method ofpersonalcontrol, and the problems of stability and steeringbecome identical. The third kind, automatic stability,is a development of the method of personal control,the mechanism which corrects a deviation being putin operation automatically instead of by the pilot.Of these three, inherent stability is obviously themost desirable if there are no countervailing dis-advantages attached to it. No attention on the partof the pilot is required and there is no mechanismto get out of order. Inherent longitudinal stabilityis easily obtained, but the problem of inherent lateralstability is more complicated.

All that is required to ensure inherent longitudinalstability is a surface placed horizontally behind theaeroplane so as to meet the air edgewise, when theaeroplane is flying in its correct position (Fig. 11, A).This surface is called the horizontal tail. Normallythe air passes over the upper and lower surfaces ofthe tail without striking it. If the head of the aero-plane tilts downward, the air strikes the upper surfaceof the tail and raises the head again (Fig. 11, B). Ifthe head tilts up, it is brought down in a similar way.This action of the tail depends both on its size and


63/184

Ill] AEROPLANE STABILITY 47on its distance behind the main planes. To increaseits effect either of these two may be increased. Ofcourse the effect of a tail may be too small to keepthe aeroplane steady : the early gliders seem to havemade the mistake of using tails which were too smalland too close to the main portion of the aero-plane. This principle, like many of those mentionedpreviously, is borrowed from birds.

BFig. 11. A. Aeroplane with neutral tail in normal flight.B. Aeroplane righted by air-pressure on tail.

A tail which meets the air edgewise may be calleda neutral or non-lifting tail. It is not absolutelynecessary that the tail should be neutral ; the essentialpoint seems to be that there should be two surfaces,of which the front one meets the air at a greaterangle than the rear one ; in other words the formersurface experiences a greater Lift in proportion toits area than the latter. Many aeroplanes have alifting tail. In some the tail is even set parallel to


64/184

48 AERIAL LOCOMOTION [OH.the main planes. Here the stability is properlyexplained by the fact that the air is cast down bythe main planes and so meets the tail at a smallerangle. In some aeroplanes the rear surface is thelarger and forms the main supporting surface, thesmaller surface in front being carried tilted up tomeet the air at a greater angle. On the whole theneutral or non-lifting type of tail seems to be best.Hargrave kites, however, and the Voisin type ofbiplanes which have been developed from them, havelarge tails set parallel to the main planes and bothpossess great stability. In all successful aeroplanesat the present time the principle of inherent longi-tudinal stability is adopted. Even the Wrights haveabandoned their early practice and now fit horizontaltails.

The ease with which the aeroplane can be turnedround must be taken into account in determining thesize and position of the horizontal tail necessary forstability. Take a rod and a weight which can beattached to the rod at different positions along it.Whirl the rod and the attached weight round in acircle. The further the weight is from the centre ofthe circle, the greater is the effort required to startthe rod whirling. Again divide the weight into twoequal parts and place them some distance apart onthe rod. It will be found that it is harder to startthe rod and weights whirling than if the two weights


65/184

in] AEROPLANE STABILITY 49are placed together midway between their formerpositions. The same principle is exemplified in theflywheel of an engine. By placing as much of theweight as possible in the rim, the difficulty of startingor stopping the wheel is increased. A body whichoffers a great resistance to being turned about anaxis is said to have a large moment of inertia aboutthat axis. On this principle in an aeroplane the lessthe difference either in height or longitudinally (i.e.behind or in front of each other) between the positionsof the different parts, the easier it is for the tail toturn the aeroplane round, and the better the stability.To ensure longitudinal stability it is best to place thedifferent parts as far as possible side by side and atthe same height. This is the arrangement adoptedin the Wright aeroplane, the pilot, passenger andengine being placed side by side. Most other con-structors however place the pilot, engine, etc. in aline from front to rear, an advantageous arrangementfrom the point of view of Head-resistance and thesaving of engine power.The problem of inherent lateral stability is very'difficult to solve. Suppose an aeroplane has acquireda sideways cant. The forces acting on it are theweight, the propeller thrust, and the air-pressure. Thislast is the same in amount before and after canting,but its direction turns round with the aeroplane. Itis clear that none of these forces will right the cant.

H. & P. 4


66/184

50 AERIAL LOCOMOTION [CH.The combined effect of the weight and the Lift, whichhaving turned with the aeroplane does not now actvertically, is to make the aeroplane move in a side-ways and downward direction towards the depressedwing-tip. Only after this sideways motion has begun,can an air-pressure be brought into play to right thecant. This fact is often urged against dependenceon inherent lateral stability. The recovery in thisway, it is said, must be slow, and the accompanyingsideways motion is objectionable as it takes theaeroplane out of its course. These are rather reasonsfor supplementing the inherent stability by otherdevices which would correct the cant more rapidly.The inherent stability should be retained for the sakeof safety and to allow the pilot to devote his attentionfor a time to other matters.

To produce inherent lateral stability the sidewisemotion set up by a cant must exert an air-pressureon some part of the aeroplane. This at once suggeststhe use of fins, or vertical surfaces, in suitable positions,meeting the air edgewise when there is no motionsideways. By an adaptation of the principle ofsuperposed planes two or more fins placed side byside at sufficient distances apart may be used insteadof one large fin. At first it seems obvious that a finplaced above the aeroplane would raise the depressedwing-tip. In such a position, or in fact in any position,it is very doubtful if one fin will ensure stability.


67/184

in] AEROPLANE STABILITY 51The action of the fin has to be sufficient to right theaeroplane and not great enough to cant it still moretowards the other side. It seems better to use twofins. One may be placed in front and one behind,either level with or slightly above the main portionof the aeroplane, certain relations being maintainedof course between the sizes and positions of the fins.Another possible combination would be one fin aboveand one behind, the fin behind being more than acertain distance from the main plane.

Instead of vertical fins other devices may be used.The fin above may be replaced by two surfaces towhich the name stabilisers may be given. Thesewould be placed respectively far to the right and leftof the aeroplane near the wing-tips, inclined upwardsfrom the centre of the aeroplane, and meeting theair edgewise in normal flight. These, in conjunctionwith a fin at a proper distance behind, should producelateral stability.

Aeroplane wings set at a dihedral angle producean effect similar to that of a fin above or a pair ofstabilisers. The wings are inclined so that their tipsare a little higher than their centre, forming thus avery much flattened V when viewed from the front orrear (see illustrationofAntoinette monoplane, page 87).

This principle, which is often used in practice,seems to be very efficient in producing lateral stabilitywhen adopted in conjunction with a fin behind. But it42


68/184

52 AERIAL LOCOMOTION [CH.is open to the serious objection that it causes a wasteof engine power. The Lift on each wing is partiallywasted, for it does not act vertically upward owing tothe inclination of the wing. To compensate for thisthe wings must be made larger; this increases theDrift, which again necessitates increased propellerthrust and increased engine power.Another device in connection with lateral stabilitywas first introduced by Mr Weiss and has latelyproved successful in the aeroplane constructed byLieutenant Dunne. Here the wings extend hori-zontally from side to side, but are twisted so thatthe Lift is greatest at the centre and diminishesgradually until, near the tips, it vanishes altogether.At the tips it is said to diminish still more, that is, itbecomes negative and pushes the wing down instead ofup. This variation of the Lift is obtained by alteringthe form of the cross-section of the wing from thefront edge to the trailing edge. Very probably thebest form for the wings of an aeroplane has not yetbeen discovered. Two things can be varied atdifferent distances along the wing the cross-sectionand the inclination of the wing to a horizontal linedrawn from one wing-tip to the other. Horizontaltwisted wings, wings set at a dihedral angle, andhorizontal wings with stabilisors fitted as extensions,are all particular ways of making these variations.Many other combinations might be devised and tested.


69/184

in] AEROPLANE STABILITY 53Whatever form of wings is used, it is probable that avertical fin in some position would also be requiredto give good lateral stability.

In Hargrave kites and the original Voisin biplanesvertical panels between the main planes and betweenthe planes of the biplane tail are utilised to givelateral stability. In kites and gliders they actsatisfactorily, but aeroplane constructors seem tohave decided that their disadvantages outweigh theiradvantages.One of the objections made to inherent lateralstability has been mentioned. Another disadvantageassociated with inherent stability, both longitudinaland lateral, is the fact that variations in the windmust cause pitching and rolling. If the wind altersits direction so as to blow more upwards or down-wards, the same effect is produced on the tail as if theaeroplane tilted. The aeroplane then must move asa ship does in pitching. If the wind veers round soas to strike the aeroplane on one side or the other,the same air-pressure is produced as if the aeroplanemoved sideways, and the inherent lateral stabilitymust make it cant or roll. With this rolling andpitching are associated the difficulty and danger offlying in a high wind. Flying in a wind, as alreadystated, is exactly the same as flying in calm air ; butthis needs the qualification that the wind referredto is perfectly steady, always blowing from the same


70/184

54 AERIAL LOCOMOTION [CH.direction at the same rate. In practice the windnever blows in this way, it always comes in gustsand also changes its direction. In addition it variesin different places, especially when diverted byobstacles or inequalities in the ground. In generalthe higher the wind, the greater its variations andthe greater the difficulty of flying in it.An aeroplane may be upset by a sudden squallas a sailing ship may be capsized, or a succession ofgusts may make it pitch or roll more and more untilat last it overturns. Ships have been lost in thisway, rolling dangerously under the influence ofcertain waves and at last overturning. The designersof the ships had not taken account of this source ofdanger ; but when the problem was investigated it wasfound easy to design ships with inherent stabilitywhich would not roll excessively under the actionof any waves they would ever meet with, and wouldright themselves even after heeling very far over. Itis possible that aeroplanes could be designed to fulfilthe same conditions, both longitudinal and lateralstability, i.e. both pitching and rolling, being takeninto account.

The study of these problems is much more com-plicated in the case of the aeroplane than in that ofthe ship. Take the problem of longitudinal stabilityand pitching. When the aeroplane pitches, the airmeets the aeroplane at a different angle so that the


71/184

in] AEROPLANE STABILITY 55Lift and the Drift must change. The alteration in theLift makes the aeroplane rise or fall, and the altera-tion in the Drift changes its velocity. In fact anyone of the three things the pitching motion, thechange in the direction of motion, and the changeof velocity must influence the other two. A com-plicated problem such as this cannot satisfactorilybe investigated without the use of rather advancedMathematics. This does not mean that the problemcan be solved by Mathematics alone, but thatMathematics is absolutely necessary in order thatexperiments may be intelligently devised and thatthe correct deductions may be drawn from them.Some well-known writers on Aviation decry Mathe-matics, but do not seem to be aware that Mathematicsis only a powerful means of reasoning and thatMathematicians, like other people, are unlikely toreach right conclusions, if they start from wrongpremises.

Important advances in the mathematical discussionof aeroplane stability were initiated by ProfessorG. H. Bryan and Mr W. E. Williams at Bangor in1903. Mr Williams has also studied the question ofstability experimentally. He photographed the actualpaths of model gliders through the air by attachingmagnesium wire to them. By placing a rotating wheelin front of the plate he was also able to measure thevelocity of the gliders. The trace of the path is


72/184

56 AERIAL LOCOMOTION [CH.divided by means of the wheel into a series ofdashes with intervening spaces ; by measuring thesedashes and gaps the velocity can be estimated.The problem of lateral stability and rolling issimilarly complicated. As explained before, a rollingor canting motion is accompanied by a motion of theaeroplane from side to side, which in its turn willgenerally cause the head of the aeroplane to veerround. Here again we have three motions influencingone another and the problem is even more complicated.The Wrights were fully aware of the danger toan aeroplane possessing inherent lateral or longi-tudinal stability of being overturned by a suddensquall. They also considered any rolling or pitchingundesirable, and for this and other reasons decidedon complete absence of inherent stability. Theyaimed at making their aeroplane uninfluenced bygusts of wind so far as pitching and rolling wereconcerned, and relied exclusively on the action ofthe pilot to restore it to its correct position aftertilting or canting. This system necessitates greatskill and consequently a long period of training onthe part of the pilot, and during flight his constantattention is demanded. These must be considereddisadvantages so far as the popular use of aeroplanesis concerned. The less the skill required on the partof the pilot and the more attention he can give toother things, his engine, for example, or the country


73/184

in] AEROPLANE STABILITY 57over which he is passing, the more popular willaeroplanes become. Even so far as flying in gustyweather is concerned, the Wright aeroplanes do notseem to have any advantage over others. TheAntoinette monoplane is generally credited with thebest performances in high winds, and it possessesinherent stability both longitudinal and lateral.Though it is impossible to say how much of thecredit for these flights is due to the design of theaeroplane and how much to the skill of the pilot,it is certain that Antoinette aeroplanes with theirinherent stability are better adapted for flight ingusty winds than Wright aeroplanes with theirabsence of stability. Recently, as stated elsewhere,the Wright brothers have departed from theirprinciples with regard to longitudinal stability.An aeroplane for general use should possessinherent longitudinal and lateral stability. Therange of stability should be as large as possible, thatis the aeroplane should right itself even after tiltingor canting far over. The means employed to givestability should not cause excessive rolling or pitchingin any winds likely to be encountered, nor shouldthey permit the overturning of the aeroplane by asudden squall. Whether all these requirements canbe satisfied is a matter for investigation. The inherentstability should be supplemented by a system ofpersonal or automatic control, especially in the case


74/184

58 AERIAL LOCOMOTION [OH.of lateral stability. The control system would beespecially useful near the ground, in the neighbour-hood of any obstacle or of other aircraft, while theinherent stability would ensure safety when highabove the ground with nothing in the vicinity.There is a third sort of stability directional,which keeps the aeroplane's head directed towardsthe same point of the compass. This sort of stabilityis inseparable from lateral. It is impossible to ensurecomplete inherent directional stability. If the headof an aeroplane has been turned by some accidenttowards a different point of the compass, no air-pressure can be exerted which will turn it back toits old direction. The lateral and directional stabilitypossessed may be such that after interference by agust of wind or some such accident, the aeroplanewill settle down into steady motion as before buttowards a different point of the compass.

2. ControlThe three different ways in which an aeroplane

can turn round have been described in connectionwith longitudinal, lateral, and directional stabilityrespectively. For purposes of steering and of bringingthe aeroplane back to its correct position after anydeviation, three controls are necessary by which thepilot can turn the aeroplane about a transverse


75/184

in] AEROPLANE CONTROL 59horizontal axis, a fore and aft axis, and a verticalaxis respectively. The three corresponding controlsmay be called longitudinal or vertical control, lateralcontrol, and directional or horizontal control.For longitudinal control a horizontal rudder,usually called an elevator, is required. This is placedeither before or behind. In the latter case it may beidentical with, or form part of the tail which giveslongitudinal stability. When not in use it meets theair edgewise, and the pilot by a system of wires andlevers, can raise or depress the rear edge so that theair strikes the upper or under surface as required.If the elevator is in front, the head is tilted down orup respectively by the air pressure on the elevator,while if the elevator is behind these directions arereversed. It must be remembered that the elevatorcannot be used alone to direct the course of theaeroplane up or down. Its use is limited to turningthe head in the required direction, and it should berestored to its normal position when this has beenaccomplished. For a short time under the influenceof its previous velocity the aeroplane will move inthe direction in which its head is turned ; consequentlythe elevator can be used to make the aeroplane takea jump. The path which will ultimately be followed,depends on the balance of the aeroplane and on themutual relations of the weight, the propelling force,the Lift and the Drift, the latter two depending on


76/184

60 AERIAL LOCOMOTION [OH.the velocity. To make the aeroplane ascend, itshead might be turned up by means of the elevator,and at the same time the engine accelerated, formore power is required to drive the aeroplane uphill. This upward motion might be obtained, withoutthe use of the elevator, by engine acceleration aloneif the stability is sufficient. If the propeller thrust isincreased, as already explained, the velocity mustincrease ; as a result the aeroplane will rise inthe air on account of the increase in the Lift. Thesame considerations apply to the descent, the pro-peller thrust being diminished instead of increased.

If the elevator is carried normally in a differentposition during flight, all the conditions of flight arealtered. The balance, the direction in which theaeroplane meets the air, the velocity required forsupport, and the propeller thrust necessary for motionin any path, are all affected. By making the mainsurfaces themselves adjustable or by detaching themand substituting others, greater alterations in theseconditions may be made.The problems of lateral and directional controlare linked up together. When one side of theaeroplane is to be raised, an increased Lift is obtainedon that side with or without a diminution of the Lifton the opposite side. In general an increase of Driftwill accompany the increase of Lift, the wing whichis raised will also be retarded, and the direction in


77/184

in] AEROPLANE CONTROL 61which the aeroplane is heading will be altered. Thedirectional control must be used to correct this.Again, when the aeroplane is flying round a curve thewing towards the outside of the curve moves fasterthan the wing on the inside, and the Lift on theformer must be greater than that on the latter. Thewing on the outside of the curve is consequentlyraised, and to correct this, or rather, as we shall see,to limit the rise to the correct amount, the lateralcontrol must be used.

Wing-warping is the mode of lateral control mostin use. The front edge of the wing remains rigid,while the rear edge near the wing-tip is pulled alittle down or up. The lowering of the rear edgeincreases the angle at which the air meets the wingand consequently increases both the Lift and Drifton it. The raising of the rear edge diminishes boththe Lift and Drift. The motions are usually controlledby a single lever or wheel. Ailerons may be usedinstead of wing-warping. These are small supple-mentary planes carried far to the right and left nearthe wing-tips. When not in use they are horizontaland meet the air edgewise ; by depressing or elevatingthe rear edge the same effect is produced as inwing-warping. Theoretically ailerons have the ad-vantage. If they are linked up so that the rear edgeof one is depressed exactly as much as the rear edgeof the other is raised, the Drifts on the respective


78/184

62 AERIAL LOCOMOTION [OH.ailerons will be equal and there will be no tendencyto turn the head of the aeroplane round. In practice, itis stated, the rudder has to be used on account of thedifficulty of adjusting the ailerons to obtain this result.For directional control a vertical rudder is placedeither before or behind the aeroplane. This is movedand acts in exactly the same way as the rudder ofa ship. It may form part of or be identical with oneof the fins used to give lateral and directional stability.Each of the three controls may be actuated by aseparate lever or wheel, one operated by each handand one by the feet. More frequently two andsometimes all three are linked up to a single lever.A lever, supposed for the sake of simplicity to bevertical, may be mounted so as to be capable of fourdifferent motions first, a backward and forwardmotion about a pivot; second, a similar movementfrom side to side ; third, a sliding movement up anddown through a collar ; and fourth, a rotation also ina collar, produced either by an attached lever like abicycle handle or by a wheel as is usual in motor-cars. Different combinations of these motions areused by different constructors. It is of course a greatadvantage if one of the pilot's hands is left free.The steering of an aeroplane to the right or leftround a curve is a complicated operation. Thesteering of a motor-car seems to be effected bymerely turning the steering wheel, but the principal

1911 aeriallocomotion00harprich

Documents