Flight, January 1936

THE D.H. DRAGONFLY
New Structural Methods Introduced in Latest De Havilland Production: Great Simplicity of Design Results in Economy of Parts: Very Complete Equipment Standardised

   BEARING a strong family resemblance to the D.H.89 Rapide, the new De Havilland differs a good deal, structurally more than aerodynamically, from that successful machine. Like the D.H.89, the D.H.90 (or Dragonfly, as it will be known) is a twin-engined biplane of familiar De Havilland plan form, with the two engines and the undercarriage faired into the lower wing. The machine has, however, been "cleaned up" in different respects as compared with the D.H.89, and is a good deal smaller in dimensions, areas and weight.
   Of the changes made in structural design, those incorporated in the wing layout are most noticeable. Whereas the D.H.89 has two sloping struts from engine mountings to top of fuselage, two inner and one outer inter-plane strut on each side, as well as a diagonal strut to brace the undercarriage laterally, the new D.H.90 has but three wing struts on each side, and the undercarriage is full cantilever. Not only is a certain amount of drag saved in this way, but the number of parts and fittings has been considerably reduced, with a resulting gain in general simplicity.
   The wing-bracing structure is interesting because of its unorthodox arrangement. There are no bracing wires in the inner bay, and those of the outer bay are in a single plane, although each lift and anti-lift wire is duplicated for safety. The double wires are, of course, placed in tandem, close together. The wing spars and ribs are of the type commonly employed in De Havilland aircraft, that is to say, of wood construction. The inner pair of inter-plane struts forms an inverted vee, meeting on a stout drag tube between the two main spars. The outer strut is a single one, also attached at its ends to the drag tube and not to either spar direct. Incidence adjustments are incorporated at the strut ends. The details are shown in the two sketches published on page 82.
   Instead of the usual wire bracing in the inner bay, the wing section of the lower wing roots has been thickened to accommodate very stout spars, which are sufficiently strong to take the loads as pure cantilevers. Several advantages arise out of this arrangement. The thick wing section of these roots makes it possible to house a petrol tank of 30 gallons capacity in each, the tanks being, of course, shaped to the wing contour. The spar ends form convenient attachments for the engine mountings and undercarriage legs, as well as for the inverted vee struts of the wing bracing. And finally, the absence of wires makes access to the cabin door easy, a step being provided on the trailing edge, and the top of the wing root forming a convenient platform.
   By supporting the landing wheels on each side, a pure cantilever undercarriage becomes possible. Each of the Turner compression struts is anchored at its top to the front spar of the wing root, light metal pads being interposed to prevent crushing of the wood of the spar. The engine bearer is a welded steel tube structure forming a cantilever beam for the engine, and having a rearward pyramid-shaped projection acting as a steadying member to the rear spar. Any tendency of the wing root to twist in a heavy landing is relieved by the inverted vee interplane strut transmitting such torque to the drag bracing of the upper wing.
   Less obvious than the changes in wing-structure design are those incorporated in the construction of the fuselage. For a good many years it has been De Havilland practice to build the fuselage as a three-ply flat-sided box, the curve being obtained afterwards by mounting the fabric covering on stringers. In the D.H.90 a totally different system has been adopted. The fuselage is, in fact, a true monocoque of wood. This has entailed a certain amount of experimentation in order to find ways and means of forming the three-ply covering to the desired shape. It is, of course, a well-known fact that a flat sheet of metal or plywood can be bent in one direction but not in two. In other words, it can be bent around a cylinder, but not around a sphere or barrel. In the D.H.90 the sides of the cabin portion of the fuselage, back to the luggage compartment, are bent to a double curvature. They are curved as seen from in front, and they are slightly curved in plan view. By experimenting, the De Havilland engineers found that if dies were used and the plywood pressed before the cement had set, it was possible to persuade the sheet to adapt itself to the double curvature. The process has now been undertaken by Saunders-Roe as a commercial proposition, and the cement used is a synthetic resin applied under heat and pressure. Aft of the luggage compartment the sides are curved in one direction only, i.e., they are straight in plan view. Consequently ordinary flat sheets can be used for the covering.
   Although the top and bottom of the fuselage of the D.H.90 are flat, the "corners" have a pronounced radius. These are built up of strips of plywood tapered off and fitted together so as to produce the desired curve, and are then glued together in the curved state. Extreme stiffness is obtained in this way, and even a considerable local pressure fails to dent these curved pieces. The only equivalents of normal longerons are light longitudinal members which serve merely to provide a surface on which the fuselage sides and curved corners can be glued.
   Regarded from the user's point of view, the De Havilland Dragonfly is a five-seater with roomy cabin accommodation and an exceptionally complete equipment supplied as standard. No attempt has been made to provide a particularly cheap aeroplane, but rather a machine which, in its standard form, is a very comfortable conveyance, equipped to go anywhere at any time, its night-landing and blind-flying equipment being absolutely complete. The cabin layout has accommodation for pilot and four passengers, one of whom sits next to the pilot, one behind him in a separate chair, and two in a sofa seat at the back of the cabin. If one attempts to judge the "commercial efficiency" according to the formula recently suggested by Mr. E. N. B. Bentley (p. 47, Flight, January 9), it is important to bear in mind not only that cabin comfort has been aimed at rather than getting the maximum number of passengers into the space available, but also that an unusually complete equipment is carried in the standard machine. This latter has naturally had its effect on the price (^2,650) Even so, however, the commercial efficiency, on Mr. Bentley's formula, works out at 3.77, which must be regarded as very good when everything is taken into consideration.
   While on the subject of figures of merit it may be mentioned that the ratio of gross to tare weight of the D.H.90 is 1.61. Superficially this indicates about an average value, but the tare weight includes 133 lb. of removable equipment, such as tool roll and grease gun, log and instruction books, airscrew covers, accumulator, three seats and cushions for sofa seat, fire extinguisher, and carpets. Furthermore, the permanent equipment is unusually complete and adds a good deal extra to the tare weight, so that everything considered the structural efficiency would appear to be above average.
   The aerodynamic figure of merit which Flight has been in the habit of using is the Everling “High-speed Figure''; for the Dragonfly this works out at 23.7, which is good for a twin-engined biplane.
   Detailed particulars of performance are given in the table on p. 80. The take-off run may appear somewhat long, but it should be realised that the permissible gross weight is rather greater than one would expect from the dimensions of the machine. That is to say, the normal gross weight figure adopted is larger than would have been quoted a few years ago. Machines are now operated in all sorts of geographic and climatic conditions, and the time appears to have come when operators must be allowed to use their discretion in the matter of load carried in their particular circumstances.
   Total tankage is provided for a still-air range of 883 miles at a cruising speed of about 125 m.p.h. The two wing-root tanks hold enough for 625 miles. Another 25 gallons can be taken in the tank built into the machine under the sofa seat.
   The standard instrument equipment of the D.H. Dragonfly includes a number of Smith's instruments, airspeed indicator, two revolution indicators, two oil-pressure gauges, eight-day watch, Husun Mark IIIA compass, one air log, P.3 type, Sperry artificial horizon and Sperry directional gyro, and three petrol meters. "Demec" navigation lights are fitted, and a Rotax landing searchlight. The electric generator is driven by a windmill, and associated with it is a 25 amp.-hr. battery. The gyro instruments are driven by two venturi tubes mounted in openings in the port lower wing root, where the airflow is of high velocity, thus enabling the blind-flying instruments to be used for the take-off. The Gipsy Major engines drive wooden airscrews with the Schwarz protective finish.

Flight, March 1936

AN AMATEUR'S "TWINITIATION''
Flying a "Little Air Liner": The New D.H. Dragonfly from the Point of View of a Single-engine Pilot
By "INDICATOR"

   DURING the past year or two I have become more and more convinced that private ownership amongst the more wealthy citizens would become universal just as soon as the right kind of aeroplane was produced. One or two manufacturers have made a genuine attempt to attract the type of owner who is really going to matter in the future, and the latest addition to the small but growing range of suitable types is the D.H.90, or Dragonfly.
   The big market is open and has been open for a number of years, but even the most optimistic of persons could hardly expect the serious business man to take kindly to the small open machine or even to the small single-engined cabin machine. Not only does he expect to go where he wants to, wet or fine, but he also expects to travel in comfort and in absolute safety with his family or his friends. Two engines are a sine qua non, as well as long range, roominess and provision for the installation of such equipment as two-way radio.
   This type of owner would probably engage a professional pilot-cum-radio operator, and the flying characteristics of the machine might not, within reason, interest him greatly. Nevertheless, it is unlikely that he would be entirely without natural curiosity and would probably feel the desire to take over during long flights. Dual control, therefore, is another essential.
   At the other end of the scale is the enthusiastic owner or would-be owner to whom the greatest interest is, or will be, in the work of piloting and navigating. Even a yacht is not much fun if a professional skipper is in permanent command. It is for this type of owner that I am speaking, and, as a single-engined light aeroplane pilot of no startling experience or skill, I can say right away that the Dragonfly would present no difficulties to any pilot with more than fifty or a hundred hours' solo flying to his credit.

Unnecessary Trepidation

   In common with the majority of club-trained amateurs, have always looked upon the medium-sized or large multi-engined machine as something rather terrifying and possessed of quaint and uncontrollable characteristics. On the various occasions when I have been left in control of devices known, euphemistically enough, to readers of the lay press as "giant air liners," I have treated them with respect and have handed over with a sigh of relief before the approach and landing. Placed precariously some yards ahead of the wings, one feels that neither a take-off nor a landing can be carried out by normal methods, and I have only comforted myself with the thought that other people manage to conduct these vessels with reasonable accuracy.
   When, therefore, Mr. Buckingham, the D.H. demonstration pilot, courageously vacated the seat of the mighty in my favour, I played nervously with the various handles of the "90" before summoning courage and pushing forward the throttles. Beyond the expected facts - a natural swing must be corrected, and, since the rudder is not directly in a slipstream, the pedals must be used with alacrity and generosity - there was nothing to trouble anyone possessed of normal common sense. When pulled off she stayed off and continued on her way at a considerable angle and with a steadily gathering speed.
   At once it was apparent that we of the light aeroplane movement have never understood the meaning of the words "forward view," nor have we realised the mentally-cramping effect of cramped surroundings. As for flying, the Dragonfly did it for herself. Within four minutes of taking over for the first time I was pushing happily through a cloud layer with eyes glued, quite unnecessarily, on the standardised Sperry equipment.
   Once above the clouds and after playing at turns for a few minutes, we - that is, Mr. Buckingham, myself and another member of the staff of Flight - played musical chairs. In other words, we climbed about the aeroplane while trying different seats. With the possible exception of two or three well-known transport aeroplanes, the "90" is the quietest machine in which I have ever flown and we conversed without a trace of that conscious raising of the voice which is usually necessary. The view even from the rear seat is good, and there is a solidity about the furnishings and furbishings which should satisfy the least airminded of citizens. The importance of this point cannot be over-stressed. After all, the Dragonfly makes a really admirable charter, machine and many taxi passengers may be having their first experience of air travel therein.
   My first approach was a frank and unvarnished failure, but, since I knew it from the very moment of closing the throttles, it could hardly be considered to be an approach within the meaning of the word. The fact was that a long, gentle turn had placed the machine thoroughly badly in relation to the Hatfield boundary, and the attempt was used for a little experiment in sideslipping. The Dragonfly slips like a Moth - and there is no need to say more.
   On a second attempt I throttled back with the full realisation that 130 m.p.h. takes a little losing with a clean aeroplane, adjusted the trim, pulled up the nose until the speed had dropped below 90 m.p.h., applied the flaps and turned towards Hatfield. We crossed the boundary at 90 m.p.h. (20 m.p.h. too fast - I was busied by the round dozen Tigers and things cavorting in the air and on the ground) yet our run had finished long before the circle was reached. This first landing was a trifle "wheely," since I was not at all sure of the height of the undercarriage, but there were none of the difficulties I had expected. I simply felt the machine on to the ground. My second landing was a full three-pointer and I knew that with another half-hour's practice the Dragonfly would be entirely "mine."
   The flaps-down gliding angle of the "90" is comparable with that of the Hornet with the air brakes "on," so that there is little chance of overshooting badly and yet no difficulty such as can be experienced with a very pronounced change of attitude during the landing process. At the same time the flaps are effective enough as brakes to allow the pilot to push the nose down, in order to steepen the approach, without gathering speed too rapidly. Although the trimming gear is of the sometimes-maligned spring-loaded elevator type, the spring mechanism is so arranged that the mechanical load is not apparent on the control when this is eased back for the hold-off and landing. In other words, one feels the air flow when it is most important that one should do.
   One little difficulty experienced is worth mentioning simply because the sequence of events shows the Dragonfly again in a foolproof light. On my second take-off the machine swung so badly that I had to throttle back and start again. Another attempt ended in the same way, despite the application of more starboard throttle and full left rudder. I was beginning to feel foolish and felt relieved when Mr. Buckingham suffered the same fate. The starboard brake was temporarily binding, and the point is, that, with no experience of the machine, I still did not get into trouble and instinctively throttled back when the swing started to develop.
   Finally, it may be remarked that the Dragonfly shows no tendency to drop a wing. A full stall, engines on or off, and flaps up or down, terminates in the gentlest fall of the nose on a perfectly level keel.

DE HAVILLAND DRAGONFLY
Two 130 h.p. Gipsy Majors

   Weight (including standard equipment) 2,500 lb.
   Disposable load 1,500 lb
   Length overall 31ft. 8in.
   Span 43ft. 0in.
   Height 9ft. 2in.
   Maximum speed 144-147 m.p.h.
   Cruising speed at 1,000ft. and 2,100 r.p.m. 127-130 m.p.h.
   Range (with 85 gallons) 885 miles
   Take-off run in 5 m.p.h. wind 205-325 yds.
   Climb to 5,000 ft 7.5 min.
   Service ceiling 15,700 ft.
   Ceiling on one engine and full load 2,100 ft.
   Gliding angle (flaps down) l in 8
   Price ?2,050.

Flight, April 1936

MODERN LIGHT AIRCRAFT REVIEWED

D.H.

<...>
   For the owner who intends to use a machine for really serious long-distance touring at all times of the year, multiplied power units are almost essential, as well as night- and blind-flying equipment, and provision for two-way radio if this is specified. The D.H. Dragonfly is a machine in the “air liner” class, which is, at the same time, almost as easy to fly as a normal single-engined aeroplane. Comfort has been studied very seriously indeed, and the noises of exhaust and airflow have been reduced to a remarkable extent.
   Generally speaking, the same tactics are used in flying this machine as in flying any other cabin craft, and with the flaps down, the approach angle is as steep as that, for instance, of the Hornet Moth. Dual control is fitted, and the instruments include a Sperry artificial horizon and directional gyro, electrically operated revolution counters and an Air Log. Special attention has been paid to the interior furnishing in order that the machine shall compare more than favourably, in this respect, with a first-class car.
   The specification of the D.H Dragonfly is as follows: Weight empty, 2,500 lb.; disposable load, 1,500 lb.; span, 43ft.; length, 31ft. 8in.; maximum speed, 144-147 m.p.h.; cruising speed, at 1,000ft., 127-130 m.p.h.; climb to 5,000ft., 7.5 min.; range, 885 miles; price, ?2,650.
<...>