It has not for obvious reasons been possible, in the design here presented, to rigidly adhere to the lines of the prototype, as in the adaptation of the design to rubber-driven model form several modifications have necessarily been introduced. It has been the aim of the writer to bring the model within the constructional capabilities of the amateur; indeed, it is hardly possible to have simplified the construction further. Now, the success of a dirigible, whether full size or model, depends primarily on the observance of the fact that an airship is lighter than air, and thus, unlike the aeroplane, does not rely on speed to obtain lift. Secondarily, the lifting power of hydrogen must be remembered, and although this varies according to temperature and the purity of it, it may be taken as a general rule that hydrogen will lift 80 lb. per 1,000 cubic feet. It is a good rule to adopt a lower figure, say 70 lb. lift per 1,000 cubic feet, to allow for discrepancies. In the design here submitted, aluminium (or what is equally as good, magnalium) tube forms the framework of the body or envelope. The general arrangement will be apparent from Figs. 178 and 179, which show the model in plan and side elevation respectively. The framework is of hexagonal cross section, the longitudinal members of which terminate at each end in a brass cap, to which they are riveted with soft brass pins. It will be necessary to anneal the tubes before bending them to impart the conical shape to each end of the frame, and this can best be effected in a weak spirit flame, care being taken to keep them on the move in it, to obviate fusing them. Where it is necessary in the construction to rivet the tubes, solder should be run over the pin to take up any play. Fig. 180 shows the method of securing the longitudinal members to the brass end-caps. The tubes are first flattened out, as at B, and then riveted to the caps. This figure also shows the method of adjusting the angle on the rudder. A piece of brass tube is soldered to the end-cap; and it should be of such a bore that the No. 18 gauge wire of which the rudder is constructed makes a bare fit through it. Two similar pieces of tube, ⅛ in. long, are soldered to the rudder, to maintain the position of it. Thirteen hexagonal cross members will be required, and each is formed from ³/₁₆ in. aluminium tube. In order that they may not become out of truth, they are cross-braced with No. 35 s.w.g. piano wire, the ends of each wire being made off in a small hole drilled through the tube. Fig. 180 will make the detail clear; each cross member is riveted to the longitudinal, and the latter is flattened out at those points where the cross member is attached.

The model is driven by four elastic motors, transmitting their power through equal gearing to the twin propellers. The four motor rods may, for preference, be hollow spars, of the same cross-section as the solid ones indicated. If such are used, they may be made by ploughing a groove in a length of wood of suitable cross-section so that it represents the letter U. Two will be required for each spar, so glued together that a hollow tube is formed. At the point where it will be necessary to pierce the spar to admit the bracing outrigger, small packing pieces of birch should be placed in the grooves previous to assembling the two halves of the spar. The spars, or, more correctly, motor rods, are suspended from the envelope framework by aluminium tube outriggers of ¼-in. diameter. Fig. 181, which shows the machine in end elevation, and Fig. 182, showing the central cross-section, indicate the form they are to take. Eight of them will be required. The angles must be cleanly and accurately formed, so that the two centres of thrust lie in the same plane. It will be found good practice, when forming both the outriggers and the cross-sectional members, to make a full-size drawing of them to use as a template. More especially is this needed in the construction of the cross-sectional members, for the purpose of ensuring that, in the operation of embracing them, they are not strained in any way so as to become out of truth.

Fig. 183 shows the joint of the motor-rod outrigger to the motor rod itself. As there shown, the tube is flattened out partly to engirdle the spar, to which it is attached by pinning and clinching.

The gearing (see Fig. 184) consists of a brass framework (bound to the ends of the motor rods) which provides bearings for the gears. The use of gears is obvious; they eliminate torque on the spars, or the tendency which a single skein would have to twist the spar. Pieces of tube are passed over the shafts to bear between the gears and the gear bearing. In order to counteract the tendency of the rubber hooks to pull out straight when the rubber is in tension, the hook ends are secured as shown at A. When it becomes necessary to detach the skeins, it is only necessary to slide the tubes along the shaft to open the hook.

In Fig. 185 the kingpost attachment is shown. The spar is mortised, the kingpost forced through, glue having previously been brushed into the slot, and a pin tapped through from the side to secure it. Birch should be used for the kingposts, and their widest cross-section should be ⅜ in. by ³/₃₂ in., tapering off towards the extremities to ⅛ in. by ¹/₁₆ in.

To the ends of each motor rod are attached small No. 22 s.w.g. piano wire hooks, to which the spar bracing is made fast. A suitable length of wire is passed through the spar, and the hooks then formed. The bracing is fastened to the kingpost by a couple of turns being taken round it.

Next, four twin hooks should be made, of the form shown in Fig. 186. Sixteen gauge wire is to be used for them, bent tightly to clip the spar ends, to which they are bound. All the hooks should be covered with valve tubing to prevent them from cutting through the rubber when this latter is in tension.

The joint of the cross members of the envelope to the longeron is given by Fig. 187, from which the bend in the longitudinal, hitherto referred to, will be clear.

To the tail of the machine a pair of superposed surfaces or elevators are fixed. These are fastened at their foremost extremities to the motor-rod outriggers, and at the rear they are supported by the two cross ribs of the rudder. Small slits are to be cut in the fabric with which the rudder framework is covered to enable any slight alteration in the angle of incidence of the elevators to be effected.

Attention may now be given to the propellers. As twin screws are used, they may be made of fairly long pitch, since their torques will be opposite and consequently balanced. Fig. 189 gives a perspective view of the propeller block marked out ready for carving, and Fig. 188 shows the screw in end elevation. The screw is first carved as a true helix, and then shaped up to the form shown in the sectional view (Fig. 182). American whitewood should be used for the blocks, or, failing this, poplar would do. Circular tin discs are pinned to each side of the propeller boss, to which the shaft is soldered. Great care is essential to ensure that both propellers are of the same weight, and that each is poised; that is, assumes an angle of 180° when balanced on a shaft. They may be finished with a coat of gold size and one of varnish. Each gear is driven by six strands of ¼-in. strip elastic, well lubricated with soft soap and glycerine.

The covering is to be yellow Japanese silk, proofed with varnish diluted with 20 per cent. of linseed oil and 10 per cent. of turpentine; two coats should be given before the fabric is applied, and two afterwards. A covering strip of fabric should be glued over all seams to make the envelope as impervious as possible. And where the outriggers pass through further pieces should be glued over, and flanged on to the tube itself, being afterwards well doped. It is well to impress here the importance of making the envelope as gasproof as possible, and although a fabric entirely impervious has yet to be invented, it is possible, with care, to reduce loss of gas by percolation to a very low figure indeed.

A Lucas cycle valve is soldered into the brass end-cap constituting the rear of the envelope, for the purpose of inflation, which is effected in the following manner: A T-piece is fixed to a cycle pump, and a pipe from the gas container to the T-piece. A further tube is connected from the remaining arm of the T-piece to the valve. To inflate the envelope, release the pressure from the container until the pump handle is forced out to its full extent, and then shut off pressure and force the pump down, thereby causing ingress of gas to the envelope. Continue thus until the envelope “swells.” Of course, this method could be adopted if hydrogen from the gas jet is used, although this does not possess the lifting capacity of hydrogen procured from the balloon manufacturers.

In conclusion, it may be pointed out that the weight of the complete model inflated should not exceed 2 lb. A view of the appearance of the model in flight is given by Fig. 190.

For those who wish to design model airships of their own it may be stated that coal-gas lifts about 35 lb. per 1,000 cub. ft., or half the weight lifted by an equal volume of hydrogen. A small model would not be very successful, as the volume, and hence the lifting power, decreases as the cube of the diameter. Thus assuming a model to be built of half the dimensions given in this chapter, the weight of it could only be ½ × ½ × ½ = ⅛ of the original model, and it would be found difficult to work to this limit. The writer would point out that little success can be expected from an airship of such small dimensions, as the following elementary calculation will show. It takes 35,000 cub. ft. of hydrogen to support 1 ton. Then 1 cub. ft. of hydrogen supports ²²⁴⁰/₃₅₀₀₀ lb. Now, the cubic contents of a model dirigible, we will assume, is

and therefore the total weight it is capable of supporting is

From this it will be seen that it is extremely improbable that a model can be built to this weight.

No more suitable covering than gold-beater’s skin exists. If the model is of the rigid type, then the covering should be stretched over the framework that imparts the ichthyoid shape to the envelope. Strips of the fabric must be attached with mucilage over all seams to make the envelope as impervious as possible. It is worthy of note that full-size airships have as many as three or four coverings to eliminate loss of hydrogen by escape through the pores of the fabric. If, however, the reader contemplates building a model of the nonrigid type, a wooden hull should be cut and the fabric fitted up to this. The hull should represent the shape of the inflated envelope.

Compressed air also lends itself to model airship propulsion.

In conclusion, a word of warning: do not place the model near any fire, gas, match, etc., as any small leaks may cause an explosion.

In building model airships, it is advisable to remember that by doubling the diameter of the envelope we get four times the capacity and hence four times the lift for only double the weight. This will be seen from the following calculation:

Lift of airship 3 ft. long 2½ in. diameter equals, as we have just seen, ·026 lb.

By doubling the diameter the calculation becomes:

From this it will be seen that if it is desired to lift a certain weight with a lighter-than-air craft, the minimum capacity should first be calculated and the diameter (which should always be as large as possible) then varied to obtain the required capacity.

Although it has been stated that the lift of hydrogen is 70 lb. per 1,000 cub. ft., it will be in order to explain two important considerations. It will be clear that, firstly, a full-size balloon must be inflated to a much higher pressure than a model, owing to the heavier mass of material to be forced out to form. As a direct adjunct to this fact it will be seen that percolation will be high. Consequently the lift weight ratio 70: 1,000 is somewhat a low estimate for a model, since, firstly, it will only require to be inflated to less than half the pressure of a full-size balloon; and, secondly, percolation will be considerably less. The writer intentionally gave the full-size limit in order that, should the builder’s model fail to come within the prescribed limit of 2 lb., its flying capability would not be appreciably impaired.

Further, the envelope is kept to its shape by a framework inside, and is consequently full of air. Now if we start pumping hydrogen in it will make the contrivance heavier instead of lighter, because no outlet is provided for the escape of air. This could easily be remedied by standing the model on its end with the cycle valve at the top, with a valve or tap on the bottom brass cap. Open this bottom tap to allow the air to escape, and allow the hydrogen to fizz in through the cycle valve. Hydrogen will not escape through the bottom tap until all the air has been displaced. If the hydrogen cannot be had at a high enough pressure to operate the valve itself a valve should not be used.

It may in such a case be necessary to fit an induction valve to the pump. The envelope should be inflated to as low a pressure as possible, otherwise the lift is correspondingly reduced. In all cases it is exceedingly difficult with model airships to prescribe any definite formula, and although the lift coefficient for models exceeds the figure stated, the conjectural nature of the purity (and hence of lift) of hydrogen in various parts of the country renders it a safe one to use for all practical purposes.