Test Planes Wiki
Advertisement


The Berlin B-9 was a powered successor to the FS17 research glider, both aircraft being used to test the prone piloting position used by the Henschel Hs 132 dive bomber.

Description[N 1][]

For the mostly two dimensional movements that are used to control an aeroplane, we usually have the pilot in a sitting position. It is a common and a natural position. It gives the pilot a great deal of freedom to guide the aeroplane. Yet, the Wright brothers used the forward reclining position for their first flying attempt.

There are several alternatives to the standard sitting position:

  1. To lay: prone on the chest, either fully stretched or in a kneeling position;
  2. To lay supine: on the back;
  3. To be seated in a tilting seat.

Lying on the back is the most favoured for high-speed manoeuvres. Unfortunately, very limited forward and downward visibility as well as placing the pilot in a psychologically vulnerable position (It is well known that animals will lie on their back as a sign of surrender), this position is not practical.

A tilting seat combines both the usual comfortable sitting position as well as a forward tilted position for higher acceleration manoeuvres. Unfortunately, a tilting seat needs extra room as well as its associated kinematic mechanism. Further, moving the position of the pilot causes mechanical and operating problems with the aircraft’s controls.

When lying prone, the pilot gains an excellent downward view, which most pilots find rather unusual.

The following table outlines the pros and cons of the positions mentioned:

Position of pilot Effective influence of acceleration in g vertical
to pilot that the pilot can cope with before losing consciousness
Sitting position, lower legs vertical Max. 6g over 3-4 seconds
Sitting position, lower legs angled forward Max. 6.5g over 3-4 seconds
Sitting position, upper body angled forward, lower legs angled forward as previously Max. 8g over 3-4 seconds
Lying supine with pilot looking toward the rear Max. 15g over 120-160 seconds
Lying prone with pilot facing the direction of flight Max. 12g over 120-180 seconds

It should however be noted that prolonged periods of very high acceleration, where the prone or supine position have a distinct advantage, are not encountered in normal flight. In normal flight, the periods of high acceleration tend to be relatively short although severe as can be seen in tables relating to centrifugal force experimentation. The position of the pilot and the correlation with his capacity for resisting g-forces is obvious.

The maximum acceleration in g-forces that the human body can withstand depend on:

  1. The degree of acceleration.
  2. The length in time of the acceleration.
  3. The direction of acceleration.
  4. The pilot’s physical condition at the moment of acceleration.

A pilot that is able to withstand high acceleration in military and practical terms has a great advantage. A fighter pilot in the normal sitting position can usually withstand 5g. A fighter pilot lying prone can withstand 12gs. In practical terms, this means that the pilot lying prone can significantly reduce the radius of his turns and pullout gradients. For example: a low flying aircraft moving at a speed of 700km/h needs approximately 700m at 5g to make a complete turn. The pilot will only be able to take this acceleration for 4 to 5 seconds. With the pilot lying prone and flying at the same speed, acceleration can be comfortably increased to 10g whilst the turn radius can be reduced to only 390m. In a dogfight the pilot flying the aeroplane with the smallest turn radius easily out manoeuvres his opponent and brings himself into a superior fighting position.

Another important advantage of the prone position is that the pilot presents the smallest possible target section so reducing the amount of armour plating or shielding to a minimum. Aircraft that have excellent visibility are highly favoured for reconnaissance, fighter and bomber tasks.

The design of the experimental aircraft Berlin B9[]

For the purpose of practically testing the position of the pilot, the Flugtechnische Fachgruppe (Aero-technical Group) Stuttgart constructed the FS17 research aircraft. The FS17 was a glider that was designed to withstand forces up to 14g. After the completion of the test program an order was given by the DVL ((Deutsche Versuchanstalt für Luftfahrt e.V. Berlin-Aldershof) (German Experimental Department for Aerospace Reg.) to the FFG Berlin ((Flugtechnische Fachgruppe)(Aero-technical Group)) to construct a powered aircraft. FFG Berlin was chosen as it possessed the necessary workshops and technicians. In the Spring of 1943 the FFG Berlin constructed the Berlin B9 to the specifications provided.

Capabilities and general specifications[]

  • The creation of an aeroplane with optimal visibility for a pilot in the prone position.
  • Stressed to accept up to 12g, positive and negative.
  • A high degree of safety with regards to the pullout gradient.
  • Very high dive acceleration to allow the aeroplane to reach high pullout accelerations.
  • Generally very good flying characteristics to allow a true judgement of the pilot position without hindering the flying characteristics.

Construction[]

The Berlin B9 is a low winged type aircraft of standard layout. It is of mixed construction and stressed to accept 22g.

Other layouts were considered. One was similar to the asymmetric Blohm & Voss BV141 and another was rear engined in similar fashion to the Göppingen Gö 9. These concepts were not progressed as too many problems were encountered.

Fuselage[]

The fuselage is constructed of steel tubing covered by timber ribbing and fabric covering. The fuselage is trapezoidal in cross-section. Its largest frame has an area of 0.67m². The fuselage diminishes in area towards the rear and finishes in the empennage.

The cockpit is covered with a 1.5m long, clear canopy that is jettisonable. The fuselage bolts to the wings at four points.

Undercarriage[]

The single leg retractable undercarriage is borrowed from the Me108. It is raised and lowered by a hand ratchet.

Empennage[]

The empennage is of simple construction. It consists of a fin with balanced rudder and elevators that is attached to the end of the fuselage. The elevators have a range of 30 per cent (27 degrees).

Wing assembly[]

The wing assembly consists of a rectangular centre section and two trapezoid like outer sections. The leading edge is square to the fuselage for half of its length. At the point where the outer panel is connected, a trail of 2 degrees is introduced. This is held for the remaining length of the wing.

The wing is constructed of two box like spars. These are situated at 20 per cent and 50 per cent through the wing from the leading edge. Dural sheets are glued to the spars for the purpose of providing attachment points for the wing to the fuselage and the engine to the wing. Solid planking to withstand the torsional forces generated by high acceleration manoeuvrers covers the area between the spars. The mounting struts for the motors are situated within the engine nacelles. The four fuel tanks are placed between the spars on either side of the motors.

The rudder and flaps mechanisms are situated just behind the last wing spar. The flaps are situated below the fuselage. They are 20 per cent of the width of the wing and can be extended to 60 degrees.

Power plants[]

Two Hirth Type HM500 motors generating 105 PS drive two fixed pitch Schäfer propellers. Fuel capacity was 95L, while oil tank capacity was 8L.

Flight controls[]

The aircraft was designed to accommodate a pilot lying in the prone position. As such it needed a flight control system that did not load up under high acceleration and needed no extra pilot training to be able to use. A fundamental change to the flight controls was out of the question. The decision was made to employ a control column instead of a control wheel.

Cockpit layout is far more important in an aircraft designed for prone operation that in an aircraft where the conventional sitting position is employed. The cockpit must be layed out in a definite right and left side pattern. Crossing hands to manipulate controls in the prone position creates significant difficulties for pilots. Blohm & Voss encountered this problem with control columns in some of their work. They also discovered that by using a small control column, the left hand could also be used to control the aircraft if the right hand was incapacitated.

In the Berlin B 9, the right hand is used to control the elevators and ailerons. It is also given the task of releasing the pilots harness and the canopy release. The left hand operates all the other controls and instruments. The feet, in the same fashion as in a conventional sitting position, operate the rudder and brakes.

In the FS17 and the first mockups of the Berlin B 9, the control column was situated centrally. In the finished Berlin B 9, the control column was located asymmetrically and approved for right-handed flight only. Even so, the left hand can be used to control the aircraft if necessary. This design change severely restricts the downward field of view to the point where ground observation is not one of the strong points of this aircraft.

The construction of the flight controls is arranged so that the movements of the control column are that of the aeroplane in the axis of roll and rotation. Although, as with the elevators, the rudder controls could be set at an angle of 60 degrees from to give a greater range of movement and higher leverage. Even so, there is no noticeable difference in control between the prone and the sitting positions.

Several pilots were tested in a mockup of the Berlin B 9 for the amount of load they were able to apply to the controls:

Controls Max. load (one hand) Max. load (2 hands) Comfortable load
Pull 25 kg 40 kg 8 kg
Push 25 kg 40 kg 8 kg
Rudder right/column right 15 kg 20 kg 5 kg
Rudder left/column left 12 kg 15 kg 3 kg

A comparison between the deflections seen at the control column of the B9 and on other conventional aeroplanes are seen in the following table:

Elevator Width: 380mm Rudder Width: 490mm
Pulling Pushing Left Right
Berlin B 9 22° 140 mm 22° 140 mm 23° 180 mm 20° 150 mm
17° 110 mm 17° 110 mm 15° 120 mm 15° 120 mm

Normally the pilot uses his ankle to operate the rudder. Only in the case of an extreme reaction is the leg used from the hip. The feet rest in pedals made to fit the pilot’s typical fur boots. They give sufficient support to the side and the back. The rudder bar, by the use of a parallel glide, can be adjusted to the pilot’s feet over a distance of 200mm. It is possible to adjust for length in flight as it is very important that the pilot is comfortable.

The pilot’s toes operate the brakes. The footrest has a cutout in the area of the toes in which the brake pedal is situated. The pedals are activated by extending the toes. If the steering is in use, the brake pedal has a small amount of freeplay before the brakes are activated.

Instrumentation and Equipment[]

The following controls, other than the emergency canopy release, are located on the left-hand side of the cockpit:

  1. Throttles.
  2. Engine instruments (fire warning, fire extinguisher, emergency pump, ignition switches).
  3. Undercarriage (select lever, ratchet).
  4. Flaps.

Experiences gained during the development of the Berlin B9 show that those controls that are not essential for safe flight can be located behind the line of the pilot’s shoulder.

The following flight and engine monitoring instruments are reflected in a mirror so as not to take up valuable cockpit space in front of the pilot:

Distance indicator, altimeter, variometer (rate of climb indicator), compass, electric turn and bank indicator, two engine tachometers, oil and fuel pressure gauges, airspeed indicator, undercarriage position indicator lights.

To assist the pilot in orienting himself, there are horizontal and inclined lines drawn on the windshield and the side windows of the canopy.

In anticipation of further development allowances were made for an ETC50 bomb rack and an experimental propeller, the MP92. For the test flight in May 1943, the aeroplane was in its original specification.

Test flights[]

The Berlin B 9 was completed in the Spring of 1943. Under the supervision of H.W. Lerche of the experimental station at Rechlin, the aeroplane made it’s first test flight.

The test flight had two purposes:

  1. To test fly and evaluate the new aeroplane.

    To explore the aeroplane’s capability and prepare it for operation as a test bed; to evaluate the strength of the aeroplane’s structure under high load conditions; to check the safe operation of and the vibration and oscillations of the power plants. Underlying the observations of these factors was the most important task; to achieve the highest performance possible from the aeroplane.
  2. To justify the adoption of the pilot’s prone position.

    Despite the many tasks involved in keeping the aeroplane flight worthy, there was a great need to immediately evaluate the lessons learned from the first test flights. The number of official visitors wishing to see the aeroplane in flight added a great deal of pressure to the pilots involved in the test program. A number of deficiencies were discovered during the test program by independent test pilots.

As of August 1943, the Berlin B 9 was presented to official Departments involved. By November 1943 thirty pilots had flown and evaluated the aeroplane. Only one accident occurred during the entire program. This occurred when a pilot made an error that may have ended up in an aborted take-off. The damage was repaired within three weeks.

Evaluation of the test flights[]

Airframe[]

The prone position of the pilot was generally indicated as being comfortable. On occasion there was a request for softer upholstery. Fatigue and tiredness was experienced in the neck (from head lifting) and shoulder muscles from moving the upper arms and the incorrect positioning of the parachute harness. Flying in a combination of winter equipment and heavy furs was noted as being tiring.

Pilots who flew the aircraft often soon adjusted to the prone position and were able to make 1 ½ hour flights without discomfort. In gliders, flights of 5 ½ hours and in motor aeroplanes flights of 1 ½ hours were entirely possible.

A chin support was considered bothersome in horizontal flight. The cockpit configuration without the chin support and the parachute on the pilot’s back was favoured most pilots. Although under high g loads a chin support was seen as being imperative. The control column was changed to become vertical and was accepted as being more comfortable by the pilots. The forces need to control the aircraft were considered as being too low. Most pilots were used to controlling much heavier aircraft. As a result, the gearing of the rudder control was changed to increase the load needed to move the rudder in flight. Several pilots took some time to get used to the feel of the rudder. No problem was encountered with the amount of force needed to operate the elevators. Cramps that developed in flight caused some difficulty to the pilots. By performing rolling exercises on the ground, leg muscles soon became accustomed to the position and cramps ceased to develop. Pilot’s legs were very sensitive to the wrong length settings for the pedals.

The whole safety equipment package, the parachute, harness and operational layout was considered very satisfactory and up to the task. For high altitude work, a special oxygen mask was needed as the breathing tube of a standard mask fouled on the chin support during the course of normal head movements.

Visibility[]

Visibility from the aeroplane is defined by:

  1. The pilot’s blind spot.
  2. The pilot’s position in the aeroplane.
  3. The pilot’s view through the canopy.
  4. The number and position of the canopy’s struts.

In the aeroplane, the pilot’s downward view is restricted by the aeroplane itself to less than 30 degrees below the horizontal. In the upward direction, his view is restricted, without moving the head, to no more than 40 degrees above the horizontal. Although the forward view within this region is unrestricted. These limitations make the prone position suitable for the following types or aircraft:

  1. Fighter aeroplanes with speeds superior to their opponents.
  2. Bombers, because of the very good view of the ground below.
  3. High speed reconnaissance aeroplanes.
  4. Aeroplanes that normally operate or attack at angles greater than 30 degrees.

The disadvantage of the prone position becomes most obvious in relatively slow aeroplanes, which normally need protection from enemy fighters and in normal fighter aircraft because of the narrow field of view above the horizon and no view to the rear.

Flight characteristics[]

The Berlin B 9 was able to achieve accelerations of 8.5g when pulling out of dives and 6g over several seconds in steep spiral climbs. Accelerations of these magnitudes are not endurable by pilots when in the normal seated position. At the beginning of the test program, these forces were only recognised by the heaviness of the head and limbs. These forces did not impair the pilot’s mental and physical reactions. Because of this, pilots often underestimated the number of gs they had pulled.

The Berlin B 9’s speed and ability to generate higher forces was restricted by the fixed pitch and relatively low rotational speed, Schwarz propellers.

Technical Data[]

Dimensions[]

  • Wingspan: 9.40m, Length max: 6.06m, Height max: 2.32m, Wheel track: 2.84m, Tyre size: 550x150mm. Tyre pressures: Medium, Wheel brakes: Hydraulic.

Surface areas[]

  • Wings and fin: 11.9m2, Rudder: 0.488m2 , Flaps: (total) 0.666m2, Wing chord; 7.45, Wing planform; Right-angle trapezoid, Dihedral: 4 degrees, Stress loading; ?22g, Depth at wing root; 1.48m , Depth, fuselage; 0.845m, Average fuselage depth; 1.266m.

Elevators[]

  • Stabiliser area: 1.365m2, Elevator area: 0.585m2, Total area: 1.95m2, Span: 3.00m.

Rudder[]

  • Fin area: 1.07m2, Rudder area: 0.63m2, Total area: 1.70m2, Height: 1.52m.

Weight[]

  • Net weight: 940kg , Payload: 175kg, Take-off weight: 1115kg.

Propellers[]

  • Type: Fixed pitch, Drive: Direct , Diameter: 2.00m, Number of blades: 2 , Rotation: Right, Swept area: 2x3.14m2.

Performance[]

  • Duration: 1 hr. 50 mins. Flight radius: 400km. Fuel consumption: 22L/100km. Max. speed: 250km/h. Cruising speed: 225km/h. Landing speed: 105km/h. Operational ceiling: 4000m. Time to 1000m: 4 min. 12 sec. Wing loading: 94kg/m2. Power to weight ratio: 5.3kg/PS. Surface area loading: 17.7PS/m2. Propeller performance: 33.4PS/m2.

Pilots who flew in the Berlin B 9[]

Pilot's Details Date
1. Eingeflogen durch Haupt-Ing. H.W. Lerche, Rechlin 10. 4. 43
2. Ing. L. Schmidt, FFG Berlin (Flugerprobung) 14. 4. 43
3. Dipl.-lng. E. G. Friedrichs, FFG Berlin und DVL (Flugerprobung) 14. 4. 43
4. Dr. med. H. Wiesehöfer, DVL 15. 6. 43
5. Ing. H. Schuhmacher, DVL 6. 7. 43
6. Dr. Ing. Doetsch, DVL 17. 7. 43
7. Prof. Kurt Tank, Focke-WuIf 30. 7. 43
8. Flugzeugführer Bartsch, Focke-WuIf 31. 7. 43
9. Flugbaumeister Mehlhorn, Focke-WuIf 31. 7. 43
10. Lt. Scheidhauer, Sonderkommando Horten 29. 8. 43
11. Flugzeugbaumeister Malz, RLM-GL/C-E2 9. 9. 43
12. Stab-Ing. Czolbe, RLM-GL/C-E2 9. 9. 43
13. Flugkapitän Rodig, Blohm & Voss 15. 9. 43
14. Flugzeugführer Rautenhaus, Blohm & Voss 15. 9. 43
15. Flugzeugführer Hilleke, Blohm & Voss 15. 9. 43
16. Stabs-Ing. Bader, Rechlin E2 23. 9. 43
17. Dipl.-Ing. Th. Goedicke, Rechlin E2 23. 9. 43
18. Stabs-Ing. Neidthard, Rechlin E2 23. 9. 43
19. Stabs-Ing. H. Böttcher, Rechlin E2 23. 9. 43
20. Stabs-Ing. Thoenes, Rechlin E2 23. 9. 43
21. Hauptmann Behrens, Rechlin E2 23. 9. 43
22. Flugkapitän Bauer, Messerschmitt 1. 10. 43
23. Flugkapitän Heini Dittmar, Messerschmitt 2. 10. 43
24. Flugkapitän Wendel, Messerschmitt 2. 10. 43
25. Dipl.-Ing. Kracht, DFS-Ainring 5. 10. 43
26. cand. Ing. Model, DFS-Ainring 6. 10. 43
27. Dipl.-Ing. Zacher, DFS-Ainring 6. 10. 43
28. Flugkapitän Zitter, DFS-Ainring 12. 10.43
29. Dipl.-lng. G. Ziegler, DFS-Hörsching 13. 10.43
30. DipI.-lng. F. W. Winter, DFS-Hörsching 13. 10.43
31. Stabs-Ing. Beauvais, Rechlin E2 27.10. 43
32. Haupt-Ing. Strobl, Rechlin E2 28. 10.43
33. Oblt. Brüning, Rechlin E2 28. 10.43

Models[]

There are two comercially available resin model kits of the Berlin B-9 - a 1/72 scale kit by Czechmaster, and a 1/48 scale kit by Lumir Vesely.

Notes[]

  1. This is from an article that appeared in Luftfahrt International (#12, Nov-Dec 1975), translated by Mr.Edmund (Eddie) & Alan Scheckenbach.[1]

Sources[]

Advertisement