Chapter II: Part 1 (1)

One of the commonest phases of interest in aviation is the desire to build a flying machine. In fact, this is very frequently the first thing the experimenter undertakes after having gone into the theory of flight to some extent. Only too often, no effort whatever is made to get beyond theory and the machine is an experiment in every sense of the word. An experience of this nature is costly—far more so than is agreeable for the student, and is likely to result in disgusting him with aviation generally. There are hundreds of schemes and principles in the art that have been tried again and again with the same dismal failure in the end. Refer to the story of the Wright Brothers and note how many things they mention having tried and rejected as worse than useless. About once in so often someone "rediscovers" some of these things and, having no facilities for properly investigating what patent attorneys term the "prior art" (everything that has gone before, from the beginning of invention, or at least patented invention) becomes possessed of the idea that he has hit upon something entirely novel and wholly original. There is no desire in the present work to discourage the seeker after new principles—undoubtedly there are many yet to be discovered. The art of flight is in its infancy and there is still a great deal to be learned about it, but there is no more discouraged inventor than he who-discovers a principle and, after having experimented with it at great expense, finds that it is only one of many things that numerous others have spent considerable money in proving fallacious, a great many years ago.

If it be your ambition to build a flying machine and you believe that you have discovered something new of value, it will be to your interest to retain a responsible patent attorney to advise you as to the prior art, before expending any money on its construction. You will find it very much more economical in the end. There are probably not more than half a dozen men alive in this country today who "know all the schemes that won’t work." The average seeker after knowledge is assuredly not likely to be one of these few, so that until he knows he is working along new and untried lines that give promise of success, it will pay him to stick to those that have proved successful in actual practice. In other words, to confine his efforts in the building line to a machine that experience has demonstrated will fly if properly constructed and, what is of equal importance, skilfully handled. Build a machine, by all means, if you have the opportunity. It represents the best possible experience. But as is pointed out under the "Art of Flying," take a few lessons from some one who knows how to fly, before risking your neck in what is to you a totally untried element. Even properly designed and constructed machines are not always ready to fly. An aeroplane needs careful inspection of every part and adjustment before it is safe to take to the air in it, and to be of any value this looking-over must be carried out by an experienced eye.

BUILDING AEROPLANE MODELS

The student may enter upon the business of building to any extent that his inclination or his financial resources or his desire to experiment may lead him. The simplest stage, of course, is that of model building and there is a great deal to be learned from the construction and flying of experimental models. This has become quite a popular pastime in the public schools and some very creditable examples of work have been turned out. The apparent limitations of these rubber-band driven models need not discourage the student, as some of the school-boy builders have succeeded in constructing models capable of flying a quarter mile in still air and their action in the air is wonderfully like the full-sized machines.

*Models with Rubber-Band Motor*. The limitations of the available power at command must be borne in mind, as the rubber-band motor is at best but a poor power plant. It is accordingly not good practice to have the spread of the main planes exceed 24 inches, though larger successful models have been built. In attempting to reproduce any of the well-known models, difficulty is often experienced in accommodating the rubber-band motor to them, as even where the necessary space is available, its weight throws the balance out entirely, and the result is a model that will not fly. This has led to the production of many original creations, but these, while excellent flyers, would not serve as models for larger machines, as of necessity they have been designed around their power plants. The rubber bands for this purpose may be purchased of any aeronautic supply house. The most practical method of mounting the motor is to attach it to the rear end of the fuselage, usually a single stick, which is accordingly made extra long for that purpose. At the other end it is attached to a bent wire fastened to the propeller in order to revolve the latter. An easy way to wand up the motor is to employ an ordinary egg beater, modified as described below, or a hand drill, inserting a small wire yoke in the jaws in place of the usual drill, or bit. This yoke is placed so as to engage the propeller blades, and the latter is then turned in the opposite direction, storing energy in the rubber band by twisting its strands tightly.

For those students who do not care to undertake an original design at the outset, or who would prefer to have the experience gained by building from a plan that has already been tried, before attempting to originate, the following description of a successful model is given. This model can not only be made for less than the models sold at three to five dollars, but is a much more efficient flyer, having frequently flown 700 feet.

_Main Frame_. The main frame of the model monoplane consists of two strips A of spruce, each 28 inches long, and measuring in cross section 1/4 by 3/8 of an inch. As shown in Fig. 1, the two strips are tied together at the front with strong thread and are then glued, the glue being spread over and between the windings of the thread, Figs. 1 and 2. The rear ends of these strips are spread apart 4 1/4 inches to form a stout triangular frame, and are tied together by cross bars of bamboo B and C which are secured to the main strips A by strong thread and glue.

_Propellers_. The propellers _D_ are two in number and are carried by the two long strips _A_. Each propeller is 5 inches in diameter, and is whittled out of a single block of white pine. The propellers have a pitch of about 10 inches. After the whittling is done they are sandpapered and coated with varnish. The thickness of the wood at the hub _E₂_, Fig. 3, of the propeller should be about 5/8 inch. At the rear ends of the strips _A_, bearing blocks _E₁_ are secured. These bearing blocks are simply small pieces of wood projecting about 5/8 inch laterally from the strips _A_. They are drilled to receive a small metal tube _T₂_ (steel, brass, or copper), through which tube the propeller shaft _T₁_ passes.

The propeller shaft itself consists of a piece of steel wire passing through the propeller hub and bent over the wood, so that it can not turn independently of the propeller. Any other expedient for causing the propeller to turn with the shaft may obviously be employed. Small metal washers _T₃_, at least three in number, are slipped over the propeller shaft so as to lie between the propeller and the bearing block.

That portion of the propeller shaft which projects forwardly through the bearing block _E₁_ is bent to form a hook _T₄_. To the hook _T₁_ rubber strips _T₂_ by which the propellers are driven, are secured. The rubber strips are nearly as long as the main strips _A_. At their forward ends they are secured to a fastening consisting of a double hook _G H_, the hook _G_ lying in a horizontal plane, the hook _H_ in a vertical plane. The hook holds the rubber strips, as shown in Figs. 1 and 4, while the hook _H_ engages a hook _T_. This hook is easily made by passing a strip of steel wire through the meeting ends of the main strips _A_, the portions projecting from each side of the strips being bent into the hooks _I_.

_Skids_. Three skids are provided, on which the model slides, one at the forward end, and two near the rear end. All are made of bamboo. As shown in Fig. 2 the front skid may be of any length that seems desirable. A 6-inch piece of bamboo will probably answer most requirements. This piece _N_ is bent in opposite directions at the ends to form arms _Z_ and _U_, The arm _Z_ is secured to the forward ends of the two strips _A_, constituting the main frame, by means of thread and glue. The strips and skid are not held together by the same thread, but the skid is attached to the two strips after they have been wound. Hence, there are two sets of windings of thread, one for the two strips _A_ themselves, and another for the skid and the strips. Strong thread and glue should be used, as before. In order to stiffen the skid, two bamboo struts _W_ will be found necessary. These are bent over at the ends to form arms _V₁_, Fig. 2. Each of the arms is secured to the under side of a strip _A_ by strong thread and glue. The arms _X_ are superimposed and tied to the bamboo skid _V_ with strong thread and glue.

The two rear skids, of which one is shown in Fig. 5, consist each of two 5-inch strips of bamboo _S_, likewise bent at either end in opposite directions to form arms _S₂_ and _S₃_, The arms _S₂_ are fastened to the strips _A_ by strong thread and glue. To stiffen the skids a strut _C₁_ is provided for each skid. Each strut consists of a 3-inch strip of bamboo bent over so as to form arms _C₂_. Strong thread and glue are employed to fasten each strut in position on the strip and the skid. In the crotch of the triangular space _B₁_, a tie bar _J_, Figs. 4 and 5, is secured by means of thread and glue. This tie bar connects the two skids, as shown in Figs. 1 and 4, and serves to stiffen them. The triangular space _B₁_ is covered with paper, preferably bamboo paper. If bamboo paper is not available, parchment or stiff light paper of some kind may be used. It does not need to be waterproof. Thus triangular fins are formed which act as stabilizing surfaces.

_Main Planes_. The main planes are two in number, but are different in size. Contrary to the practice followed in large man-carrying monoplanes, the front supporting surface is comparatively small in area and the rear supporting surface comparatively large. These supporting surfaces _L_ and _P_ are shown in detail in Figs. 6 and 7. It has been found that a surface of considerable area is required at the rear of the machine to support it, hence, the discrepancy in size. Although the two supporting surfaces differ in size, they are made in exactly the same manner, each consisting of a thin longitudinal piece of spruce _R_, to which cross pieces of bamboo _Q_ are attached. In the smaller plane, Fig. 7, all the cross pieces are of the same size. In the larger plane, Fig. 6, the outer strips _S_ are somewhat shorter than the others. Their length is 2 1/2 inches, whereas the length of the strips _Q_ is 3 1/2 inches. In order to allow for the more gradual tapering of the plane, around the outer ends of the longitudinal strips _R_ and the ribs _Q_ a strip of bamboo is tied. The frame, composed of the longitudinal strip and cross strips, is then covered with bamboo paper, parchment paper, or any other style light paper, which is glued in place.

The forward or smaller plane has a spread of 8 1/2 inches and a depth of 3 1/4 inches. The main plane has a spread of 20 inches and a depth of 3 1/2 inches at the widest portion. The author has made experiments which lead him to believe that the tapering form given to the outer edge of the plane improves both the stability and endurance of the machine.

The planes are slightly arched, although it will be found that flat planes will also give good results. The rear edge of the main plane should be placed 4 1/4 inches distant from the forward edge of the propeller block _E₁_.

The front plane must have a slight angle of incidence, just how much depends upon the weight of the machine, the manner in which it is made, and various other factors. This angle of incidence is obtained by resting the front portion of the plane on two small blocks N, Figs. 1 and 2, which are fastened to the top of the main strip _A_ by strong thread and glue.

The height of the blocks N should be about 1/4 inch, although this will necessarily vary with the machine. The blocks should be placed approximately 4 inches from the forward end of the machine. The front end of the forward plane should be elevated about 1/4 inch above the rear end, which rests directly on the main strips.

Both the front and rear planes _L_ and _P_ are removably lashed to the frame by means of ordinary rubber bands, which may be obtained at any stationery store. These rubber bands are lettered _M_ in Fig. 1.

_Winding the Rubber Strips_. The rubber strips can be most conveniently wound up by means of an egg beater, slightly changed for the purpose. Fig. 8. The beater and the frame in which it is carried are entirely removed, leaving only the main rod _E_, which is cut off at the lower end so that the total length is not more than 2 or 3 inches. The two brass strips _D_ on either side of the rod, which are attached to the pinion _Q_ meshing with the large driving wheel _H_, are likewise retained. A washer _F_ is soldered to the rod near its upper end, so as to limit the motion of the small pinion and the brass strips _D_ attached to the pinion. Next a wire _B_ is bent in the form of a loop, through which loop the central rod passes. The ends of the wire are soldered to the side strips _D_. Lastly, a piece of wire _C_ is bent and soldered to the lower ends of the side strips. In order to wind up a rubber strip, the strip is detached from the forward end of the model, and the hook _A_ slipped over the wire _C_. The opposite end of the rubber band is held in any convenient manner. Naturally the two strips must be wound in opposite directions, so that the two propellers will turn in opposite directions. By stretching the rubber while it is being wound, more revolutions can be obtained. It is not safe to have the propeller revolve more than 700 times. The ratio of the gears of the egg-beater winder can be figured out so that the requisite number of twists can be given to the rubber bands for that particular number of revolutions.

_Model with Gasoline Motor_. The next and somewhat more ambitious stage is the building of a power-driven model, which has been made possible by the manufacture of miniature gasoline motors and propellers for this purpose. Motors of this kind, weighing but a few pounds and capable of developing 1/4 horse-power or more, may be had complete with an 18-inch aluminum propeller and accessories for about $45. As is the case with the rubber-band driven model, the monoplane is the simplest type to construct, and the dimensions and details of an aeroplane of this type are given here. It will be found that a liberal-sized machine is required to support even such a small motor. The planes, Fig. 9, have a spread of 7 feet 8 inches from tip to tip, each wing measuring 3 1/2 feet by a chord of 15 inches. They are supported on a front and rear wing spar of spruce, 1/2 by 3/8 inch in section, while the ribs in both the main plane and the rear stabilizing plane measure 1/8 by 1/2 inch in cross section. There are eight of these spruce ribs in the main plane, and they are separately heated and curved over a Bunsen burner, or over a gas stove, which is the same tiling. They are then nailed to the wing spars 6 inches apart. The main spars of the fuselage are 7 feet long and they are made of 1/2 by 3/8 inch spruce, the struts being placed 1 1/2 feet apart, measuring from the rear, with several intermediate struts to brace the engine bed. Instead of using strut sockets for the fuselage, which would increase the cost of construction unnecessarily, a simple combination of a three-way wire fastener and a wire nail may be resorted to. The shape of these fasteners is shown at _A_ in Fig. 9. They may be cut out of old cracker boxes or tin cans (sheet iron) with a pair of shears, the holes in the ends being made either with a small drill or by driving a wire nail through the metal placed on a board, and filing the burrs off smooth. A central hole must also be made for the 1 1/2 inch wire nail which is driven through the main spar and the fastener then slipped over it. As indicated, this nail also serves to hold the strut. A drop of solder will serve to attach the fastener to the nail. The front of the fuselage is 9 inches square, tapering down to 6 inches at the rear. The height of the camber of the main planes is 1 1/2 inches and the angle of incidence is 7 degrees, measured with relation to the fuselage. The non-lifting tail plane at the rear which is to give the machine longitudinal stability, measures 4 feet in span by 14 inches in depth.

The running gear or front landing frame is made of 1/2 inch square spruce, all joints being made with 1/16 by 1 inch bolts. Aluminum sleeves, procurable at an aeronautic supply house, are employed for the attachment of the rubber springs and the radius rods running down to the wheels, which may also be purchased ready to install. Old bicycle wheels will serve the purpose admirably. Light steel tubes 1/2 inch in diameter are used to run these aluminum sleeves on. Two other steel tubes are joined to the lower corner of the frame by flattening them at the ends and drilling with a small hole for a nail. These are run diagonally up to the fuselage and serve as buffers to take the shocks of landing. For bracing the wings, two similar tubes are fastened to form a pyramid on top of the main plane just back of the engine. From these, guys are run to the wings as shown. The engine bed is made of 1/2 by 3/4-inch white pine, and to make it solid it is carried as far back as the rear edge of the main plane. The batteries and coil are directly attached to this plane, care being taken in their placing to preserve the balance of the machine. The rudder measures 14 inches square and is made of 3/8-inch square spruce, reinforced with tin at the joints, as it is necessary to make the frame perfectly rigid. Both sides are covered with fabric. In this case a 1-horse-power motor furnishes the necessary energy and it is fitted with an 18-inch aluminum propeller which it is capable of turning at 2,400 r.p.m. The carbureter and gas tank are made integral, and the gasoline and oil are both placed in this tank in the proportion of about four parts to one, in order to save the weight of an extra tank for oil.

Flights of half a mile are possible with this model in calm weather, but a great deal of measuring and testing of the fuel is necessary in order to regulate the flight, and "grass-cutting" should be practiced by the builder in order to properly regulate the machine. Trials have shown that the flat non-lifting tail on the fuselage gives excellent longitudinal stability, the machine rising nicely and making its descent very easy angle, so that it is seldom damaged by violent collisions in landing.

BUILDING A GLIDER

The building of hand- or power-driven models does not suffice to give that personal experience that most students are desirous of obtaining. The best method of securing this is to build a glider and practice with it. Any flying machine without a motor is a glider and the latter is the basis of the successful aeroplane. In the building of an aeroplane the first thing constructed is the glider, _i.e._ the frame, main planes, stabilizing planes, elevators, rudders, etc. It is only by the installation of motive power that it becomes a flying machine. The biplane will be found the most satisfactory type of glider as it is more compact and therefore more easily handled, which is of great importance for practicing in a wind. The generally accepted rule is that 152 square feet of surface will sustain the weight of the average man, about 170 pounds, and it will be apparent that the length of the glider will have to be greater if this surface is to be in the form of a single plane than if the same amount is obtained by incorporating it in two planes—the biplane. A glider with a span of 20 feet and a chord of 4 feet will have a surface of 152 square feet. So far as learning to balance and guide the machine are concerned, this may be mastered more readily in a small glider than in a large one, so that there is no advantage in exceeding these dimensions—in fact, rather the reverse, as the larger construction would be correspondingly more difficult to handle. The materials necessary consist of a supply of spruce, linen shoe thread, metal sockets, piano wire, turnbuckles, glue, and closely-woven, light cotton fabric for the covering of the planes.

*Main Frame*. The main frame or box cell is made of four horizontal beams of spruce 20 feet long and 1 1/2 by 3/4 inch in section. They must be straight-grained and perfectly free from knots or other defects. If it be impossible to obtain single pieces of this length, they may be either spliced or the glider may be built in three sections, consisting of a central section 8 feet long, and two end sections each 6 feet in length, this form of construction also making the glider much easier to dismantle and stow in a small space. In this case, the ends of the beams of each end section are made to project beyond the fabric for 10 inches and are slipped into tubes bolted to corresponding projections of the central section. These tubes are drilled with three holes each and bolts are passed through these holes and corresponding holes in the projecting ends after they have been fitted into the tubes, and drawn up tightly with two nuts on each bolt to prevent shaking loose. Ordinary 3/16-inch stove bolts will serve very nicely for this purpose. The upper and lower planes forming the box cell, are held apart by 12 struts, 4 feet long by 7/8 inch diameter, preferably of rounded or oval form with the small edge forward to minimize the head resistance. It is only necessary to space these equally, starting from both ends; this will bring the splices of the demountable sections in the center of the square on either side of the central section. The main ribs are 3 feet long by 1 1/4- by 1/2-inch section and their placing should coincide with the position of the struts. Between these main ribs are placed 41 small ribs, equally spaced and consisting of pieces 4 feet long by 1/2 inch square. These, as well as all the other pieces, should have the sharp edges of the square rounded off with sand paper. The ribs should have a camber of 2 inches in their length and the simplest method of giving them this is to take a piece of plank, draw the desired curve on it, and then nail blocks on both sides of this curve, forming a simple mould. The rib pieces should then be steamed, bent into this mould, and allowed to dry, when they will be found to have permanently assumed the desired curvature. Meanwhile, all the other pieces may be shellaced and allowed to dry.

_Assembling the Planes_. To assemble the glider, the beams are laid out on a floor, spaced the exact distance apart, _i.e._, 3 feet, and exactly parallel—in the demountable plan, each section is assembled independently. The main ribs are then glued in place and allowed to set, after which they are strongly bound in place with the linen thread, and the various layers of thread given a coating of hot glue as they are put on. This method is not arbitrary, but it is simple and gives the lightest form of construction. If desired, tie-plates, clamps, or any other light method of fastening may be employed. This also applies to the ribs. They are assembled by placing them flush with the front beam and allowing them to extend back a foot beyond the rear beam, arched side up in every case. They may be glued and bound with thread, held by clamps, or nailed or screwed into place, care being taken to first start a hole in the beam with an awl and to dip the nails in soft soap to prevent splitting the wood. Twenty-one ribs, spaced one foot apart, are used in the upper plane, and 20 in the lower, owing to the space left for the operator in the latter. For fastening the two planes together, whether as a whole or in sectional units, 24 aluminum sockets will be required. These may be purchased either ready to fit, or an effective substitute made by sawing short lengths of steel tubing, slitting them with the hack saw an inch from the bottom, and then flattening out and drilling the right-angle flanges thus formed to take screws for attaching the sockets to the beams. In case these sockets are bought, they will be provided with eye bolts for the guy wires; if homemade, they may have extra holes drilled in the edges of the flanges for this purpose or some simple wire fastener such as that described in connection with the power-driven model may be used, heavier metal, however, being employed to make them. The sockets should all be screwed to the beams at the proper points and then the struts should be forced into them. The next move is to "tie" the frame together with guy wires. No. 12 piano wire being employed for this purpose. Each rectangle is trussed by running diagonal guy wires from each corner to its opposite. To pull these wires taut, a turnbuckle should be inserted in each and after the wire has been pulled as tightly as possible by hand, it should be wound upon itself to make a good strong joint, as shown at _B_, Fig. 6. A fastening as shown at _A_ will pull out under comparatively little strain and is not safe. As is the case with most of the other fittings, these turnbuckles may be bought or made at home, the simple bicycle type of turnbuckle mentioned in connection with "Building a Curtiss," being admirably adapted to this purpose. In fact, the construction of the latter will be found to cover the requirements of the glider, except that the ribs are simpler and lighter, as already described, and no provision for the engine or similar details is necessary. All the guy wires must be tightened until they are rigid, and the proper degree of tension for them may be simply determined in the following manner:

After the entire frame is wired, place each end of it on a saw horse so as to lift it two or three feet clear of the floor. Stand in the opening of the central section, as if about to take a glide, and by grasping the forward central struts, raise yourself from the floor so as to bring your entire weight upon them. If properly put together the frame will be rigid and unyielding, but should it sag even slightly, the guy wires must be uniformly tightened until even the faintest perceptible tendency to give under the weight is overcome.

_Stretching the Fabric_. The method of attaching the fabric will be determined by whether the glider is to be one piece or sectional, and the expense for this important item of material may be as little or as much as the builder wishes to make it. Some employ rubberized silk, others special aeronautic fabrics, but for the purposes of the amateur, ordinary muslin of good quality, treated with a coat of light varnish after it is in place, will be found to serve all purposes. The cloth should be cut into 4-foot strips, glued to the front horizontal beams, stretched back tightly, and tacked to both the rear horizontal beams and to the ribs. Tacks should also supplement the glue on the forward beams and the upholstery style should be used to prevent tearing through the cloth. In case the glider is built in sections, the abutting edges of the cloth will have to be reinforced by turning it over and stitching down a strip one inch wide, and it will make this edge stronger if an extra strip of loose fabric be inserted under the turn before sewing it down. Eyelets must then be made along these edges and the different sections tightly laced together when assembling the glider. It is also desirable to place a strip of cloth or light felt along the beams under the tacks to prevent the cloth from tearing out under the pressure.

To form a more comfortable support for the operator, two arm pieces of spruce, 3 feet by 1 inch by 1 3/4 inches, should be bolted to the front and rear beams about 14 inches apart over the central opening left in the lower plane. These will be more convenient than holding on to the struts for support, as it will not be necessary to spread the arms so much and there will be more freedom for manipulating the weight to control the glider in flight. In using the struts, it is customary to grasp them with the hands, while with the arm pieces, as the name implies, the operator places his arms over them, one of the strips coming under each armpit. After the fabric has been given a coat of varnish on the upper side and allowed to dry, the glider is ready for use. The cost of the material should be about $30 to $40, depending upon the extent to which the builder has relied upon his own ingenuity in fashioning the necessary fittings—in any case, it will be less than the amount required for the purchase of the engine alone for a power-driven model.

*Glider with Rudder and Elevator*. It will be noted that this is the simplest possible form of glider in that it is not even provided with a rudder, but for the beginning of his gliding education the novice will not require this, as first attempts should be confined to glides over level ground in moderate, steady wind currents and at a modest elevation. Some of the best gliding flights made by Herring, Chanute’s co-worker, were in a rudderless glider. After having mastered the rudiments of the art, the student may go as far as the dictates of his ambition impel him in the direction of improvements in his glider, by adding a rudder, elevator, and warping control. In fact, it is not necessary to confine himself to the simple design of glider here outlined at all. He may take either the Wright or Curtiss machines as a model and build a complete glider, following the dimensions and general methods of construction here given, though these may also be improved upon by the man handy with tools, bearing in mind that the object to be achieved is the minimum weight consistent with the maximum strength.

*Learning to Glide*. The first trials should be made on level ground and the would-be aviator should be assisted by two companions to help him in getting under way. The operator takes a position in the center rectangle, back far enough to tilt up slightly the forward edges of the planes. A start and run forward is made at a moderate pace, the keepers carrying the weight of the glider and overcoming its head resistance by running forward at the same speed. As the glider cuts into the air, the wind caused by running will catch under the uplifted edges of the curved planes and will buoy it up, causing it to rise in the air taking the operator with it. This rise will be probably only sufficient to lift him clear of the ground a foot or two. Now he projects his legs slightly forward so as to shift the center of gravity a trifle and bring the edges of the glider on an exact level, parallel with the ground. This, with the momentum acquired at the start, will keep the glider moving forward for some distance. When the weight of the operator is slightly back of the center of gravity, the leading edges of the planes are tilted up somewhat, increasing the angle of incidence and in consequence the pressure under the planes, causing the glider to rise, and if the glide is being made into a wind, as should always be the case, quite a height may be reached as the result of this energy. Once it ceases, the tendency to a forward and upward movement is lost, and it is to prolong this as much as possible that the operator shifts the center of gravity to bring the machine on an even keel, or where at a little height, slightly below this, giving it a negative angle of incidence, which permits him to coast down the air until sufficient speed is acquired to reverse the angle of incidence and again rise so as to provide a "hill" for another coast, thus prolonging the flight considerably. To put it in the simplest language, when the operator moves backward, shifting the center of gravity to the rear, the planes are tilted so that they catch or "scoop up" the advancing air and rise upon it, whereas when he moves forward and the planes tilt downward, this air is "spilled" out behind and no longer acts as a support, and the glider coasts, either until the ground is reached or enough momentum is gained to again mount upon the wind. A comparatively few flights will suffice to make the student proficient in the control of his apparatus by his body movements, not only as concerns the elevating and depressing of the planes to ascend or descend, corresponding to the use of the elevator on a power machine, but also actual steering, which is accomplished by lateral movement to the left or right.

Stable equilibrium is one of the chief essentials to successful flight and this can not be maintained in an uncertain, gusty wind, especially by the novice. The beginner should certainly not attempt a glide unless the conditions are right. These are a clear, level space without obstructions such as trees, and a steady wind not exceeding 12 miles per hour. When a reasonable amount of proficiency has been attained in the handling of the glider over level ground, the field of practice may be changed to some gentle slope. In starting from this, it will be found easier to keep the glider afloat, but the experience at first will prove startling to the amateur, for as the glider sails away from the top of the slope, the distance between him and the ground increases so rapidly that he will imagine himself at a tremendous height, but by preserving the balance and otherwise manipulating his weight in the manner taught by the practice over the level, a nice flight of much greater distance will be made and the machine will gradually settle down to the ground much farther away from the starting place than was possible in the earlier trials, this being one of the great advantages of starting from an elevation. There is nothing that will fit the beginner so well for the actual handling of a power machine as a thorough course of gliding flights, and it is recommended that those who build gliders become proficient in their use before attempting to pilot an aeroplane, whether of their own make or not.

A further step in advance is the actual building of a full-fledged power machine, and for those who desire a simple and comparatively inexpensive type, requiring very little work that can not be performed in the home workshop, a description of the construction of a Curtiss biplane is given, while for those who are more ambitious and also have greater financial resources, the details of the building of a Bleriot monoplane are given.

BUILDING A CURTISS BIPLANE

Cost. First of all, the prospective builder will want to know the cost. The best answer to this is that the machine will cost all its builder can afford to spend upon it and probably a little more, as the man to whom the expense is not of vital consideration will doubtless not undertake its construction. Speaking generally, and there can be nothing very definite about it, in view of the great difference in the conditions, an expenditure of three to four hundred dollars will cover the complete outlay for everything but the motor. If the builder has the time and facilities for doing all the work himself, this amount may be reduced very materially. On the other hand, if he finds it necessary to purchase most of the material in form ready to assemble, it may exceed this. But it will be a great aid to many to know that there is practically nothing about the modern aeroplane which can not be found in stock at one of the aeronautic supply houses. This makes it possible for many to undertake the construction of a machine to whom it would not be feasible, or at least not an attractive project in view of the time involved, were it necessary to make every part at home. So far as becoming involved in any legal difficulties is concerned owing to existing patents, the student need not worry himself about this in attempting the construction of a Curtiss biplane, so long as he restricts the use of his machine to experimental purposes and does not try to compete with the patentees in their own field—that of exhibiting and selling machines.

*General Specifications*. Just how long it will take to complete such a machine will depend very largely upon the skill of the builder and the extent of his resources for, as already mentioned, the expense may be cut down by making all the necessary parts at home, but it will naturally be at the sacrifice of a great deal of time. For instance, the oval struts and beams may be bought already shaped from the local planing mill, or they may be shaved down from the rough by hand. Turnbuckles can be made from bicycle spokes and nipples and strips of sheet steel, or they can be bought at 12 to 15 cents each. As a hundred or more of them are needed, their cost is quite a substantial item.

Aeroplane construction doubtless impresses the average observer as being something shrouded in considerable mystery—something about which there is no little secrecy. Quite the contrary is the case in reality. Any man who is fairly proficient as a carpenter and knows how to use the more common machinist’s tools, such as taps and dies, drills, hacksaw, and the like, will find no difficulty in constructing the machine of which the details are given here. Having completed its building, he will have to draw upon his capital to supply the motor. One capable of developing 25 to 30 horse-power at 1,000 to 1,200 r.p.m. will give the machine considerable speed, as it will be recalled that Curtiss made a number of his first flights with a 25-horse-power motor. As to the weight, the lighter the better, but 400 pounds for the complete power plant will not be excessive. The machine can sustain itself in the air with less power than that mentioned, but with a heavy, low-power motor it will be sluggish in action. This is an advantage for the amateur, rather than otherwise, as it will provide him with an aeroplane that will not be apt to get away from him during his first trials, thus making it safer to learn on.

The Curtiss biplane has a spread of 30 feet, the main planes or wings being divided into sections of a length equal to the distance between struts, Figs. 11 and 12. There are five of these sections, each measuring six feet. The struts can be taken out and the sections laid flat on each other for storage. The framework for the front and rear rudders can also be jointed, if desired, making it possible to store the machine in small compass. The longest parts of the machine, when taken apart, are the two diagonal beams running from the front wheel back to the engine bed, and the skid. The horizontal front rudder is packed intact. The vertical rear rudder is unhung and laid flat on the tail. Two men can take the machine apart in a few hours, and can reassemble it in a day. Whether these particular features of construction are covered by patents can not be said, as Curtiss has declined to commit himself regarding any rights he may have to them.

_Ribs_. Two distinct types of ribs are used, main ribs and small ribs, both of the same curvature, Fig. 13. The main ribs are used between pairs of struts, to hold apart the front and rear beams; they are heavy enough to be quite rigid. Three to four small ribs are laid across each section of the planes, between the pairs of main ribs, to give the cloth the proper curvature, and to maintain it in the form desired. The main ribs are built up of six 1/4-inch laminations of wood 7/8 inch wide and securely glued together. The small ribs are made of three layers 1/2 inch wide.

The first part of the actual construction will be the making of these laminated ribs, but before describing this detail, the question of suitable material should be well considered. Both weight and strength must be figured on and this limits the choice to a few kinds of wood. Of these _spruce_ and _elm_ are the best available, with the occasional use of _ash_ to give greater rigidity. Spruce is, of course, the first choice. This wood was once considered as having no great strength, but a series of careful tests shows this belief to be unfounded. With the exception of the bed, or support for the motor and a few other parts, the Wright machines are constructed wholly of spruce.

Table I gives results of tests made with spruce from Washington and Oregon, and with elm from Michigan and Indiana. Testing scales were employed, the pieces being supported at their ends with the load in the center.

These tests were made with clear wood in each case, as knots naturally decrease the strength of a piece greatly, this depending on their size and location.

Before proceeding with the ribs themselves, the press for giving them the proper curvature must be made. Take a good piece of oak, ash, or other solid wood, 8 inches wide by 5 feet long, and dressed all over. On the side of the piece lay out the curve, the dimensions of which are illustrated in Fig. 14. First, rule the horizontal, or chord line, on it, marking off 4 feet 6 inches on this line, equidistant from each end. Then divide the chord into 6-inch sections and, at the point of each 6-inch section, erect perpendiculars beginning at the rear, 3/4 inch, 1 3/8 inches, 2 inches, and so on, as indicated on the drawing. The upper ends of these perpendiculars will form locating points for the curve. Through them draw a smooth curve as shown, continuing it down through the chord at each end. Take the piece with the curve thus marked on it to the local planing, sash and blind, or sawmill—any plant equipped with a band saw—and have it cut apart along the curve. This will cost little or nothing—acquaintance will obtain it as a favor, and acquaintance with any wood-working concern in the aeroplane builder’s home town will be of great aid. Failing this aid, the operation may be carried out with a hand saw (rip), but the job will not be as neat and will have to be cleaned up with a draw knife and sand paper, taking care to preserve the outline of the curve as drawn. As the rib press is really a mould or pattern from which all the ribs are to be bent to a uniform curvature, care must be taken in its construction.

To clamp the two halves of the press together, a dozen machine bolts will be required; they should measure 3/4 X 15 inches. If obtainable, eye bolts will be found more convenient as they may be turned up with but one wrench and a bar. The steel straps are 3/8 by 1 1/2 by 10 inches long with 3/4-inch holes drilled 9 inches apart to centers, to enclose the 8-inch pieces.

Obtain a sufficient supply of boards of reasonably clear spruce, 1/4 inch thick, 6 to 7 inches wide, and at least 4 feet 9 inches long (dressed both sides), to make all the ribs necessary both small and large. This material should be purchased from the mill as it is out of the question to attempt to cut the ribs from larger sizes by hand. Buy several pounds of good cabinet makers’ glue and a water-jacketed gluepot. This glue comes in sheets and in numerous grades—a good quality should be used, costing from 40 to 50 cents a pound if bought in a large city. Laminating the ribs in this manner and gluing them together is not only the quickest and easiest method of giving them the proper curve, being much superior to steam bending, but is also stronger when well done, as the quality of the material can be watched more closely.

Start with the making of the small ribs; apply the glue thin and piping hot in a generous layer to three boards with a good-sized flat paint or varnish brush. Omit on the upper surface of third board and apply between three others, Fig. 13. This will give two series of three each in the press. Tighten up the end bolts first, as the upper part of the press near the top of the curve is likely to be weak unless liberally proportioned. Then turn down the nuts on the other bolts. Do not attempt to turn any one of them as far as it will go the first time, but tighten each one a little at a time, thus gradually making the compression over the whole surface as nearly uniform as possible. This should be continued until the glue will no longer ooze out from between the boards, indicating that they are in close contact. Twenty-four hours should be allowed for drying, and when taken out the cracks between the boards should be almost invisible in the finished ribs.

Have the laminated boards cut by a power rip saw at the planing mill, to the dimensions shown in the drawing, making an allowance of 1/4 inch for the width of the saw blade at each cut in calculating the number of ribs which can be cut from each board. In addition, a margin should be allowed at each side, as it is impractical to get all the thin boards squarely in line. For the main ribs, apply the glue between all six boards, clamp and dry in the same manner. Thirty small ribs will be required, if three are used in each section, and forty if four are specified, while twelve main ribs will be needed for standard construction, and sixteen if the quick-demountable plan referred to is followed. It is advisable to make several extra ribs of each kind in addition. If the builder has not sufficient faith in spruce alone, despite the figures given in Table I, one of the laminations, preferably the center, or if two be employed, the outer ones, may be of ash, though this will add considerably to the weight.

To prevent the ribs from splitting open at the ends, they are protected by light steel ferrules, shown in Fig. 15. When received in the rough-sawed condition from the mill, the ribs must be tapered at the ends with a plane or spoke shave to fit these ferrules, and the sharp edges should be rounded off. In doing this, it must be remembered that the upper surface of the small ribs gives the curvature to the cloth surface, so that any tapering must be done on the lower side. The main ribs may be tapered from both sides, as it is the center line, or crack between the third and fourth laminations, that determines the curve. Every inch along this line A-inch holes are to be drilled for the lacing, Fig. 15.

The ferrules for the front ends of the small ribs are light 1/2-inch seamless steel tubing; they may be flattened to the proper shape in a vise without heating and are drilled with a 1/8-inch hole. They are driven tight on to the tapered ends of the ribs and fastened in place with a small screw. The rear-end ferrules are 1/2-inch lengths of 3/8-inch tubing, driven on and drilled with a 1/32-inch hole for the rear-edge wire. The rear ferrules of the main ribs may be the same 1/2-inch tubing used for the front of the small ribs; they should be cut off so that their ends will come in the same line as the holes in the ends of the small ribs. If the quick-demountable plan be followed, the second main rib from each end may be left long and drilled with a hole like the small ribs. The front ferrules of the main ribs should be 3/4-inch tubing of heavier gauge, drilled with a 1/4-inch hole. The finished ribs are sandpapered smooth and shellaced or coated with spar varnish. The latter is much more expensive and slower in drying but has the great advantage of being weather-proof and will protect the glue cracks from moisture. The ferrules may be painted with black enamel.

_Struts_. Before going into the detail of the construction of the remainder of the _main cell_ and its attached framing, a brief description of its parts and their relation to one another will make matters clearer. The upright struts, Fig. 15, which hold the two planes apart, fit at each end into sockets, which are simply metal cups with bolts projecting through their ends. Fig. 16. Those at the bottom of the front row of struts pass through the eyes of the turnbuckles and connections for the wire trussing, then through the flattened ferrules of the main ribs, and finally through the beam, all being clamped together with a nut. Those at the top go through the turnbuckles first, then through the beam, and finally the rib ferrule. The bolts at the back row of struts must go through the full thickness of the main ribs, and so must be longer. The drawings. Figs. 15 and 16, show the method of attachment of both the main and the small ribs and illustrate a neat method of attaching the turnbuckles—instead of being strung on the socket bolt one after another, they are riveted to the corners of a steel plate which alone is clamped under the socket.