Page:EB1922 - Volume 30.djvu/64

34

The earliest steps in England, or indeed anywhere, to unify such standards were taken by the Royal Aircraft Factory at Farnborough in 1913. They were extended and improved as experience developed under the Aircraft Inspection Depart- ment (A.I.D.) in England (towards the end of 1915), and later under the British Engineering Standards Association, which in 1921 was instrumental in founding in Paris the " Comite Inter- national pour 1'Unification Aeronautique " to internationalize the same work.

Components. Fuselages, wings, undercarriages, tail planes and controlling surfaces, prior to 1914 were not, save in one or two cases, designed as self-contained units, i.e. their manufacture was usually completed during erection into the aeroplane. This involved hand- fitting, trial and error adjustment, constant inspection and slow production, while spares were not interchangeable.

By 1915 each component became a unit in itself, made to limits, corresponding with the connexion points, and interchangeability was safeguarded by the use of jigs and fixtures. By 1919 even compo- nents were subdivided into standardized parts, and the assembly of components into a complete aeroplane could be effected after deliv- ery to the field. The jigs and fixtures were usually confined to the location of junction fittings on which the structure was erected. These replaced the fixtures of 1915, which held all members of the component in position during construction, but proved not to be satisfactory, owing to the distortion of the finished piece on re- moval from the fixture.

Girder types of construction, such as fuselages, wings, etc., were latterly constructed to jigs rather than on fixtures, in order that their truth of erection might be more permanent. Monocoque con- structions, however, were always built on cradles or moulds, which definitely determined their final shape; the individual members, being free from initial load, were free from distortion on removal from the mould.

The development of portable gauges (gauge points mounted on tensioned wires) occurred in 1916.

In 1917 component junctions were designed so that all positioning was determined by one joint, clearance in one direction being allowed on the remaining joints; the gauging of components was simplified thereby, and many of the more costly gauges could thus be super- seded by simpler ones used in conjunction with a measuring operation. Woodwork. Wood is eminently suitable for light construction and for obtaining a rapid output by machining. The mechanical properties and suitability of various timbers were little known in 1913. Bamboo, the lightest timber, was found unsuitable in about 191 1 ; it lacks uniformity in size, and is difficult to connect at the end of members. Ash (Fraxinus excelsior) and hickory (Carya alba, Hichora ovatd) were early used, but hickory is scarce, and variable in its mechanical properties, and ash is heavy as well. Ash is re- stricted to use where high flexibility and shock-resisting are essential. Silver spruce (Picea Sitchenis, Carr.) was introduced in 1913 for spars, struts, longerons and other members, being uniform, light and suitable for machining for weight reduction.

Between 1913 and 1915 accurate information of the strength and elasticity of this timber was acquired. Methods of converting the timber for the various uses were determined in order to eliminate de- fects peculiar to coniferous timbers, such as spiral grain, cross and diag- onal grain, dote or rot, gum pockets, alternating hard and soft grains, low density, wide-ringed timber and brittle or lifeless timber (brash). The great demand in 1916 in England led to the importation of unseasoned timber, needing to be conditioned for use. The French and Americans had already experience of this. Kilns were erected in England (on the Sturtevent system of drying). Humidities, tem- peratures and time periods of drying were determined. Control of the moisture-content of timber was found to be essential.

The larger aeroplanes in 1916 and 1917, and the demand in excess of supply Tor best of spruce of long lengths, led to spars being made of short lengths joined together, the joints being situated at points of low stress. A study of various joints in 1918 led to the adoption

FIG. 1 8.

of the plain vertical scarf joint with an inclination of I in 9, reinforced by bolts through the splice, and bound with fabric (see fig. 18). Shorter timbers glued together as laminations then became permis- sible for all spars, and defects could thus either be cut out or re- inforced. Joints in these laminations, after being admitted for a period, were ruled out in 1919.

To supplement the supplies of silver spruce in 1917 the following

timbers were tried in 1918, the peculiarities of each being allowed for:

Quebec Spruce (Picea alba and Picea nigra, Link )

White Sea White Deal (Picea excelsa, Link.).

White Sea Red Sea Yellow Deal (Pinus sylvestris, Link.).

West Virginia Spruce (Picea rubeus, Sargent).

North Carolina Spruce (when this is the same as West Virginia Spruce, but grown in North Carolina).

Louisiana Red Cypress (Bald.).

Port Orford Cedar (Chamaecyparis Lawsoniana, Murr.).

New Zealand Kauri (Agathis [Dammara] auslralis).

Canadian White Pine (Pinus Strobus, Link.).

Oregon Pine (Psettdotsuga Douglasii, Carr.).

Cypress, which is very variable, liable to brittleness and unsuitable for glueing, was barred in 1918. Oregon pine, which is liable to frac- ture under shock, and may split when cut into small dimensions must be restricted to struts and used in the solid. Small knots in the deals can be allowed in laminations if the knots l>e distributed to obtain uniformity of the member. Laminated struts were used in 1919, with fabric binding to safeguard against the opening out of joints. Early in 1918 box sections, which have all the advantages of laminating, were used, and their use continues.

About 1915-6 the glues used in the above processes were classified into three grades: (i) the best for airscrews; (2) for less highly stressed joints; (3) for unimportant details. Glue shops were main- tained at a constant 70 Fahrenheit. Micro-investigation of glued joints proved the value of carefully preparing the timber and glue; timber was aged to prevent warping, by storing in the 7O-F. rooms for long periods before glueing. Roughing of the surfaces to be glued was adopted to secure keying.

In 1915-6 it was found that if an entire series of laminations were glued in one operation before clamping the first joint would become chilled before the clamping occurred. Later, by using trained crews and special appliances for quick glueing and clamping, the en bloc process of glueing with the more rapid output became possible and satisfactory. Where heated-glue rooms could not be used, " liquid " glue or jelly glues (containing an ingredient which delays the setting poin^of the glue, thus allowing of ordinary temperatures 55 F. to 60 F. with ample time for assembly of parts) were adopted.

Metal Fittings. In 1910 fittings for the structures, attachment of bracings, etc., were made of mild steel, a metal selected, no doubt because it could be worked cold. This was often used in double thickness to ensure against flaws. Oxy-acetylene welding was often used in joints, even in some that were subject to stress. Tubes and plates were welded together to make sockets, and bent to shape with- out being subsequently normalized. Failures at welds led to the substitution in 1915 of mild-steel drop forgings. These were ma- chined all over to save weight and to get the size accurate to toler- ances too small for the stamping industry at that time.

The correct temperature for forging and subsequent heat treat- ment of the forging in high tensile steel was not currently known. The facilities were lacking, and the control of the temperatures needed was left too much to the estimate of the skilled operative. Stampings brittle and unreliable for use, as well as difficult to ma- chine, were made. In 1915-6 the impact test, long known but little used, was supported by the War Engineering Committee of the Royal Society, and was found valuable for ensuring that the material so tested would bear prolonged shock stress.

By 1917 the call for speedy output led to a reversion from forgings to sheet-metal sockets and fittings, using a low carbon sheet-steel of 26 tons' ultimate tensile strength. The pressings were shaped in jigs which ensured an adequate radius at the bend, and they were nor- malized to remove strains due to bending or punching. Where com- plicated fittings were built up of simpler pressings these were riveted and soldered together to avoid welding. Dip-brazing of such constructions came in in 1918, with the advantage that the temperatures could be better controlled than when brazed with a slow-pine. Such pressings are interchangeable and need less gaug- ing and inspection. Turnbuckles, universal joints, shackles, etc., litherto machined from the bar, were re-designed for quicker manu- 'acture from sheet metal.

Bracings. In 1910-1 8o-ton steel " piano wire " was much used r or bracing the structure, but the fastenings for this had only some 60% of the strength of the wire; the loops stretched, and the struc- ture was soon distorted. Flexible cables spliced on to wiring plates and adjusted by turnbuckles were then used with greater safety, but these also stretched and increased the air resistance, to reduce which wooden fairings were applied to the cables. Solid wires swaged to streamline form, and left thick at the ends for screwing, were made as early as 1911, but they were difficult to manufacture. In 1913 this
 * air section was abandoned for the elliptical, to allow of rolling instead

of swaging the rods, while a special steel and heat treatment evolved by the Royal Aircraft Factory overcame the difficulties. These wires were not generally adopted till, in 1915, standardized aeroplanes ed to a demand which warranted bulk production.

Wires of streamline section were swaged, not rolled, because these asymmetrical sections tend to curve over sideways as they pass out 'rom the rolls. The elliptical-section wires were called " Rafwires," o distinguish them when they were standardized. The screwing of he end of these wires was carried on after heat treatment (at 550 "".). Subsequently the wires were tempered at a lower temperature