The Horten Tailless Aircraft
by K.G. Wilkinson, B.Sc. D.I.C.

Horten VIII & IX

3.10  Horten VIII


     This was to have been a flying model of a proposed six-engined trans-Atlantic passenger transport weighing 100,000 kg.  The span was to be 40 m with an aspect ratio of 10 and sweepback of 28°.  Power units were six Argus AS 10 C engines.
     To make the aircraft attractive to R.L.M. and thus get backing for the project, the Hortens added a rear loading cargo carrying body with an internal space approximately 14’ x 10’ x 6’; this was not part of the design for the full size aircraft.  With construction under way, another modification was made (but not disclosed to R.L.M.).  This consisted of removing the nose of the cargo body, replacing the nose wheel by wheels on either side of the body and putting a large venturi tube with a 2m x 2.7m throat inside to form a flying wind tunnel.  They expected to get about 500 mph airspeed in the throat combined with low turbulence – this they proposed to check by the sphere drag method.  Later they hoped to be able to test models of their aircraft which could be made of wood because of the absence of dust in the airstream.
     Construction was proceeding at Gottingen and was 50% complete at the cessation of hostilities.  The steel tube framework for the venturi center section was finished.

Estimated Weight and Performance Figures

Max. all up weight as a wind tunnel                9,000 kg
Max. all up weight as a cargo carrier 
     Without takeoff assistance                         15,000 kg
     With rocket assisted takeoff                      20,000 kg

At 23,000 kg the sea level rate of climb at full power would be zero.
At 9,000 kg rate of climb at 180 kph was expected to be 6.5 – 7 m/sec.

Estimated trimmed CLmax’s were
               No Flaps             1.4
            With Flaps             1.6
     CL for Takeoff             1.1

Aerodynamic Design

     The design of the wing and controls was similar to that of the Horten IV.  Washout was large, 7°, to give trim without elevator deflection at cruising CL.  Elevons were the three stage type with 35% Frise nose on the outer flap, and 22% on the middle and inner flaps.  Compensating geared tabs which could also be used a longitudinal trimmers were fitted to the inner flaps.  Maximum control deflections were a follows:

(Note:  All figures in degrees)
.. -------
-------    ---------
Stick fwd. & central
Stick back & central
-10  or -15
-10 or -15
Stick central & to port
Stick central & to stbd.

     Trailing edge split flaps with a constant chord of 80 cm were to be fitted between the engines.
     Drag rudders were of the H VII “trafficator” type with vent hole balance plus spring centering.  Projection was about 1 meter.
     Wing sections are shown in Fig. 18.  Root thickness is about 16%, with the usual reflexed center-line, graded to an 8% symmetrical tip section.


     Wing structure was in seven parts; a welded steel center section with pilot and co-pilots seat and three outer wooden wing panels per side.  The wooden structure was of single spar D-tube form with subsidiary trailing edge ribs.
     At the factory in Gottingen the center section was found in a semi-complete state, D-noses for the inboard wing panels were finished and spars and ply noses for the outer panels were under construction.  Much of the work on components such as engine bearers, petrol systems, undercarriage etc., had been completed and the six engines were in crates at the works, with one spare.  Unfortunately all drawings had been taken and many of them seem to have been buried by Horten employees near Kilenburg, in the Russian sector.


     The fixed main wheels were arranged in tandem pairs on either side of the fuselage and took 85% of the static weight of the aircraft.  The castering nose wheel was retractable on the cargo version and had to be mounted on a stalky strut because of the high wing layout.  Static ground incidence was 2.5°.

3.11  Horten IX


     The H IX was a single seat fighter bomber of 16 m span with twin jet engines, being a further development of the H V and H VII designs.  Fig. 19 is a general arrangement drawing made from a wooden model found at Gottingen, where the first two of the type were built.
     Four aircraft of the H IX type were started, designated V.1 to V.4.  V.1 was the prototype, designed as a single seater with twin B.M.W. 003 jets, which were not ready when the airframe was finished.  It was accordingly completed as a glider (Fig. 20) (not reproducible) and extensively test flown.  D.V.L. instrumented it for special directional damping tests to determine its suitability as a gun platform.  V.2 was completed (also at Gottingen) with two Juno 004 units and did 2-hours flying before crashing during a single engine landing.  The pilot (Ziller) apparently landed short after misjudging  his approach.  V.3 was being built by Gotha at Friedrichsrodal as a prototype of the series production version.  V.4 did not get beyond the project stage but was to be a two-seater night fighter with an extended nose to house the extra man (Fig. 19) (missing).
     In shape, the H IX was a pure wing with increased chord at the center to give sufficient thickness to house the pilot and the jet units, which were placed close together on either side.

Aerodynamic Design

     The H IX started as a private venture and the Hortens were very anxious to avoid failure so they avoided aerodynamic experiments wherever possible.  A lower sweepback was used than on the H V and H VII and laminar flow wing sections were avoided as a potential source of trouble.  Wing section at the junction with the center sections was 14% thick with maximum thickness at 30% and 1.8% zero Cmo camber line.  At the centerline thickness was increased locally to 16% to house the crew.  The tip section was symmetrical and 8% thick.  Horten also believed that since the compressibility cosine correction to drag was based on the sweepback of the maximum thickness line, the ordinary section would show little disadvantage.
     Wing twist was fixed by consideration of the critical Mach number of the underside of the tip section at top speed.  This gave a maximum washout of 1.8°.  Having fixed this, the CG was located to give trim at CL = 0.3 with elevons neutral.  In deciding twist for high speed aircraft, CD values were considered in relation to local CL at operational top speed and altitude (10 km in the case of the H IX).  Twist was arranged to give minimum overall drag consistent with trim requirements.  The wing planform was designed to give a stall commencing at 0.3 to 0.4 of the semi-span.


     Wing structure comprised a main spar and one auxiliary spar or wooden construction with ply covering.  The center section was built up from welded steel tube.  Wing tips were all metal.  The undercarriage was completely retractable and of tricycle type the front wheel folding backwards and the main wheels inwards.  The nose wheel was castering and centered with a roller cam.  When resting on the ground, wing incidence was 7° and the nose wheel took about 40% of the total weight.

Engine Installation

     The jet engines were installed at -2° to the root chord and exhausted on the upper surface of the wing at 70% back from the nose (Fig. 22a & 22b).  To protect the wings the surface was covered with metal plates aft of the jet pipe and cold air bled from the lower surface of the wing by a forward facing duct and introduced between the jet and the wing surface.  The installation angle was such that in high speed flight the jest were parallel to the direction of flight.

Control System

     Lateral and longitudinal control was by single stage elevon control flap with 25% Frise nose and compensating geared tap balance.  (This system was also used on the H VII, see para. 4.6.)  The pilots control column was fitted with a variable hinge point gadget, and by shifting the whole stick up about 2” the mechanical advantage could be doubled on the elevons for high-speed flight.
     Directional control was by drag rudders.  These were in two sections, slight movements of the rudder bar opening the small (outboard) section and giving sufficient control for high speed.  At low speeds when courser control was necessary the large movement also opened the second spoiler, which started moving when the small one was fully open.  By pressing both feet at once, both sets of spoilers could be operated simultaneously; this was stated to be a good method of steadying the aircraft on a target when aiming guns.  The Hortens stated that the spoilers caused no buffeting and claimed an operating force of 1 kg for full rudder, with very little variation in speed.  The operating mechanism is illustrated in Fig. 28.  A change was made from the original H VII parallel link system to improve the control force characteristics.  With the new system, aerodynamic forces could be closely balanced by correct venting of the spoiler web, leading the main control load to be supplied by a spring.  The cover plate of the spoilers was spring loaded (Fig. 27) to form an effective seal with the rudders closed; this device was used on most Horten spoiler and dive brake designs.
     On further models of the H IX it was proposed to fit the “trafficator” type rudder tried experimentally on the H VII.
     Landing flaps consisted of plain trailing edge flaps (in four sections) on the wings, with a 3% chord lower surface spoiler running right across the center section, which functioned as a glide path control.  The outer pair of plain flaps lowered 27° and the inner pair 30° – 35° on the glider version V.1.  On V.2 mechanical trouble prevented the inner pair operating and all flying was done with the outer pair only.  The center section spoiler could be used as a high speed brake and gave 1/3 g at 950 kph.  No dive recovery flap was considered necessary.


     Proper performance tests were not done on V.2 before its crash and top speed figures were calculated values, checked by Messerschmitts.  The following figures were remembered by Reimar Horten:


All Up Weight, Including Ammunition and Armor           8,500 kg (18,700 lbs.)
All Up Weight, Excluding Ammunition and Armor          7,500 kg
Wing Area                                                                                      52 sq.m (566 sq.ft.)
Wing Loading                                                                                 33 lb./sq.ft.
Fuel (I2 Crude Oil)                                                                   2,000 kg (4,400 lbs.)

Performance at 7,500 kg (16,500 lbs.)

Takeoff Run                                                                          500 m
Takeoff Speed (10° Flap)                                                  150 kph (95 mph)
       (Note:  This corresponds to a CL of 1.30 which is the stated stalling CL of the aircraft.)
Top Speed (at Sea Level)                                                 950 kph (590 mph)
(CDo estimated to be 0.011)

Calculated ceiling was 16 km (52,000’).  Engines would not work above 12 km as the burners went out.

Rate of Climb at Sea Level                                              22 m/sec (4,300 ft/min)
(Note:  This has been checked roughly by observation.)

     In tests against the Me 262 speeds of 650-700 kph (400-430 mph) were obtained on about 2/3 throttle opening.  This appears to be the only flight test figure available.
     Messerschmitt sent performance calculators to the Horten works to check their estimates.  The method suggested by D.V.L. for getting the sweepback correction to compressibility drag was to take an area of 0.3 x the root chord squared at the center section as having no correction applied, and then apply full cosine correction over the outer wing.  Sweepback angle was defined as that of the quarter chord locus.  Test data was available for CDv. for zero sweepback.
     The Messerschmitt method was to base sweepback on the max t/c locus and to scale Mach number by the square root cos Ø.

Stability and Control

     The H IX V.1 was flown by Walter Horten, Scheidhauer and Ziller.  Scheidhauer did most of the flying (30 hours) at Oranienberg, Horten and Ziller flew for about 10 hours.
     D.V.L. instrumented the aircraft for drag and directional stability measurements.  No drag results were obtained because of trouble with the instrument installation – apparently an incidence measuring pole was fitted which could be lowered in flight and glide path angle was obtained  from the difference between attitude and incidence measurements.  One day they landed without retracting the pole.  Directional oscillation tests were completed successfully and an advance report was issued (10 pages of typescript) by Pinsker and Lugner fo D.V.L.
     The essence of the results was that the lateral oscillation was of abnormally long period – about 8 sec. At 250 kph and damped out in about 5 cycles.  At low speeds the oscillation was of “dutch roll” type but at high speed very little banking occurred.  Many fierce arguments took place at D.V.L. on desirable directional stability characteristics , the Hortens naturally joining the “long period” school of thought.  They claimed that the long period would enable the pilot to damp out any directional swing with rudder and keep perfectly steady for shooting.  It was found that by using both drag rudders simultaneously when aiming, the aircraft could be kept very steady with high damping of any residual oscillation.
     Lateral control was apparently quite good with very little adverse yaw.
     Longitudinal control and stability was more like a conventional aircraft than any of the preceding Horten types and there was complete absence of the longitudinal "wiggle" usually produced by flying through gusts.  Tuft tests were done to check the stall but the photographs were not good enough for much to be learned.  Handling was said to be good at the stall, the aircraft sinking on an even keel.  There seems to be some doubt, however, as to whether a full stall had ever taken place since full tests with varying CG and yaw had not been done.  Although the stick was pulled hard back, the CG may have been too far forward to give a genuine stall.
     Directional stability was said by Scheidhauer to be very good, as good as a normal aircraft.  He did not discuss this statement in detail as he was obviously very hazy about what he meant by good stability and could give very little precise information about the type and period of the motion compared with normal aircraft.
     Scheidhauer had flown the Me 163 as a glider and was obviously very impressed with it; he was confident enough to do rolls and loops on his first flight.  We asked him how the H IX V.1 compared with the 163; he was reluctant to give an answer and said the two were not comparable because of the difference in size.  He finally admitted that he preferred the 163 which was more maneuverable, and a delight to fly (he called it “spielzeug”).
     The H IX V.2 with jet engines was flown only by Ziller and completed about 2 hours flying before its crash.  This occurred after an engine failure – the pilot undershot, tried to stretch the glide and stalled.  One wing must have dropped, for the aircraft went in sideways and Ziller was killed.  Before the crash a demonstration had been given against an Me 262; Horten said the H IX proved faster and more maneuverable, with a steeper and faster climb.
     In spite of the crash, Horten thought the single engine performance satisfactory and said the close spacing of the jets made single engined flying relatively simple.

..PREVIOUS SECTION................................NEXT SECTION

Back Home...