Accueil
Les planeurs de F. Budig

La machine de Budig (en version motorisée), à la Wasserkuppe (en 1923) - [1]

Année du premier vol : 1921 ?
Pays : Allemagne
Développement :

Friedrich Wilhelm BUDIG

Caractéristiques
Tryptique
Historique

Références

Vers page maquettes

CARACTÉRISTIQUES
  Type : planeur puis planeur motorisé Atelier de fabrication : --  
 

Envergure : -- ,-- m

Vitesse minimale : --- km/h  
  Longueur : -,-- m Vitesse maximale : --- km/h  
Hauteur : -,-- m Taux de chute mini : -- m/s à -- km/h
  Allongement : --,-- Finesse max : -- à -- km/h  
  Surface alaire : -- m² Profil d'aile : ?  
  Charge alaire : -,- kg/m² Nombre de sièges : 1  
  Masse à vide : -- kg Nombre de machines construites : ?  
  Masse maximale : --- kg Années de fabrication : ---- - ----  
  Ballast : non Techniques de construction : Bois et toile  

Pas de spécifications connues, relatives à ce planeur.

TRYPTIQUE ET PLANS
 


Extrait du brevet, un dessin en vue de côté d'une machine proposée par F. Budig. [3]

Les photos et le croquis ci-contre permettent de se faire une idée assez précise de la formule aérodynamique de cette machine. C'est un sesquiplan : en effet, l'aile inférieure est de plus faibles envergure et corde que l'aile principale.Le plan canard stabilisateur à l'avant, en deux parties : la partie principale servant à la stabilisation longitudinale automatique, et derrière cette dernière, un volet classique de commande de profondeur, actionné par le pilote.

Sur les deux versions, il y a également un plan horizontal à l'arrière, de très petite surface, qui semble être fixe (?).

On note que le dessin ci-contre prévoit une hélice tractrice, alors que la machine qu'il a construit ensuite est à hélice propulsive.

 

HISTOIRE

Parmi les machines inscrites au concours de la Rhön, en Allemagne, en 1921, figurait un curieux planeur de type canard, dessiné et construit par Friedrich W. Budig.
Le plus étonnant dans cette machine n'était pas sa géométrie, mais la mise en oeuvre d'un dispositif de contrôle automatique de la stabilité en tangage de l'appareil,
particulièrement original. Les photos ci-dessous attestent des essais effectués, mais les performances en vol ne semblent pas avoir été convaincantes,
car les comptes-rendus de ne font guère état de cet appareil.


 Les réactions du planeur dans le vent sont d'abord testées sans pilote
Les résultats ont dû être positifs puisque qu'ensuite F. Budig fit d'autres essais
de sa machine, la pilotant lui-même -
[2]
Puis presque deux ans plus tard, au début de l'année 1923, la revue anglaise Flight International publie un article qui reparle de la machine de F. Budig, dans une nouvelle version, avec un fuselage profondément modifié dans sa partie arrière, afin de pouvoir y installer un moteur. Il s'agissait d'un moteur de moto flat-twin BMW développant une puissance de 4 CV. [1]

Version motorisée du Budig [1]
Peu d'informations ont été données sur les performances obtenues avec ce planeur motorisé, et il semble manifeste que la puissance développée par le moteur n'était pas suffisante pour permettre un décollage de la machine sans assistance, nécessitant un départ traditionnel à la catapulte. Une fois en l'air la machine devait être capable de voler à l'horizontale, ou au mieux de s'élever avec un faible taux de montée. L'hélice étant montée en prise directe sur l'axe moteur, son rendement était certainement très faible.
Comme de plus la géométrie de la machine induit une traînée importante, il ne fait pas de doute qu'il ne devait pas y avoir beaucoup de surplus de puissance pour permettre une montée.  En fait la motorisation devait seulement permettre de prolonger quelque peu le vol.

Un moteur ayant équipé la machine de Budig est actuellement exposé au Deutsches Museum Flugwerft Schleissheim. Budig aurait donc assez rapidement remplacé le BMW initial par ce Salmson A.D. 3. Ce moteur, fabriqué par la Société de Moteurs Salmson, Billancourt, France. Ce moteur d'un peu moins de 1000 cc de cylindrée, développe une puissance de 12 CV (18 kW) à 1800 tours/minute. Il pèse 34 kilogrammes.

Ce moteur a fait ses preuves dans les années 1920 en gagnant le Grand Prix de la Moto-Aviette et le Prix Solex (en 1925).


Salmson A.D. 3 au Deutsches Museum Flugwerft Schleissheim [photo ClaudeL]

Le moteur et la curieuse hélice montée en direct sur l'arbre moteur. On peut avoir de sérieux doutes quant au rendement d'une telle hélice. [1]



En mai 1924 se tint le deuxième concours de vol sans moteur de Rossiten, près de 
Koenigsberg (Prusse occidentale). Ce concours a fait date dans l'histoire du vol à voile car il vit le record du monde de durée porté à 8 heures 42 minutes 9 secondes par Ferdinand Schultz avec son FS-3 Besenstiel (le 11 mai 1924). Hentzen et Budig participaient à ce concours et tous deux annoncèrent qu'ils avaient l'intention de se procurer un moteur Blackburne pour équiper leurs planeurs respectifs.

Et il semble que les machine de Budig ne firent plus parler d'elles...

 Quoi qu'il en soit, le Budig peut très probablement être considéré comme l'un des tout premiers planeurs motorisés, avec le Max de Hentzen (Wampyr motorisé) qui lui est contemporain.

DESCRIPTION DU SYSTÈME AUTOMATIQUE
Qu'en était-il du système de stabilisation longitudinale automatique de Budig ? On ne peut pas espérer trouver description plus précise que celle du brevet lui-même déposé par Budig en 1921. [3]
Son  texte est volontairement donné ici sans traduction, afin d'apprécier le style propre aux rédactions de brevet !


" F.W. BUDIG
Flying machine
Application filed June 16, 1921

1,419,447                                Patented June 13, 1922

To all whom it may concern :
Be it known that I, FRIEDRICH WILHELM BUDIG, citizen of the state of Prussia, residing at Falkenberg-Grunau, near Berlin, in the state of Prussia, Germany, have invented certain
new and useful Improvements in Flying Machines (for which I have filed applications in France, Aug. 1, 1914, and in Germany Jan. 29, 1920 and Nov. 20, 1920), of which the following
is a specification.
This invention relates to a steadying device. It consists in the combination of the supporting planes proper with an automatically adjustable steadying plane. This latter plane is
rotatably supported upon an axle supported in its turn by the flying machine, and its forms, together with an immovable surface with which it is connected, preferably by bellows,
a hollow space within which a depression may be produced in known manner by the mediation of a feeler surface.  The steadying plane is adjusted to various inclinations by that depression, contrary to the action of a spring. Besides these arrangements, other ones also forming parts of the present invention relate to means for correctly adjusting
the steadying plane, as is particularly necessary for soaring flying apparatus.
Reffering to the accompanying drawing, Figure 1 is a vertical section through
a motorless flying machine provided with the novel device in question, this latter
being shown in the position it occupies at too slow flight ; Figure 2 is a side-view
of the machine, only some upper parts (2, 5, 14) and a lower part (9) being shown
in section, and the position of the automatic steadying being that which
it occupies at normal speed ; Figure 3 is a diagrammatical illustration of the
succession of the swing motions of the machine if flying in soaring manner ;
and Figure 4 is a side-view of a soaring flying machine having a small auxiliary motor.
 The upper body 2 of the steadying plane is hinged to a horizontal axle 1 firmly
supported by the frame of the flying machine. Affixed to said body 2 is
 a counterpoise 3, the arrangement being such that the total centre of gravity
 of the poise and the body 2 is located above the common axis of rotation
so that in the case of changes of the speed of the flying machine the law of inertia
initiates a favourable swinging or turn of the upper body 2.
Parallel with the body 2, and having the same breath as this body, is an immovable plate 4, to the rear edge of which is hinged the horizontal rudder 5 for altitude steering.
The body 2 and the plate 4 are connected with each other by bellows 6 forming or enclosing a chamber 7 which communicates by a channel or passage 8 with the hollow
 feeler plane 9 which has the shape of an inverted supporting plane and is provided with a slot 10 extending over the whole of its breadth. 11 is a spring which tends to turn
the body upwards about the axle 1, that is to say, to expand the bellows. The position of the parts 2 and 6, if no partial vacuum within the bellows counteracts the spring,
is that shown in Figure 1.
The bellows communicates with the atmosphere not only through the channel 8, but also through an aperture 12 which is at the front of the machine and constantly open.
The sectional area of the aperture 12 is smaller than that of the slot 10. The supply of the air through the aperture 12 does not materially affect the degree of rarefaction
in the bellows as long as the air
entering through that aperture flows off through said slot, as in the case during flying with normal speed. But if the speed decreases, the body 2 can rise,
under the pressure of the spring 11, far more quickly
 than would be the case if no supply of air through the aperture 12 would take place, because, as is known by experience, air, even at reduced speed,
can enter only with great difficulty through the slot 10 into the bellows.
In order to accelerate the swinging movement of the body 2 in the direction to the plate 4
at an increase of the speed of flight, the space 7 is connected by one or more air-channels
13 provided in rigid parts of the structure preferably tubes, and forming a kind of girder
with the spaces of hollow bodies 14 united with the main supporting planes.
The tubes 13, or their equivalents, carry the above described combination and arrangement
of parts and serve also for fastening the means provides for the landing.
The hollow body 14 forms a rigid piece for and of the supporting plane and its front portion
 the blowing-at edge of said plane. The supporting wall of that body 14 differs from an
ordinary hollow supporting plane chiefly by a slot 15 which is provided in the lower surface
of the projecting front part of the body 14, similarly to the slot 10, and the length of
which is about one fourth of the length of the slot 10, provided, that the machine
has an aperture such as 12.
By measurements it is known that at that place of the supporting frame profile where
 the slot 15 is located a strong depression is produced at quick flight.
This observation confirms that at slow flightthe slot 15assists the action of the aperture 12,
and at quick flight assists the action of the slot 10.
Thus, if the angle of incidence at which the flight takes place is reduced, then, by the addition of the slot length 15 with the slot length10, a certain definite amount of air
to be sucked out of the bellows space 7
 is exhausted more quickly and the body 2 moves correspondingly quicker in the direction of the plate4 ;
 that could not be the case without the slot 15.
The manner in which the body 2 is turned in the one or the other direction will become clear from a description dealing with flying
in soaring manner, as follows :

Supposing the machine be soaring motionless in a constant head-wind under a large angle of incidence. In this case the lifting power R acts at the main supporting plane,
and the lifting power X acts at the steadying surface. The lifting power of the horizontal rudder 5 which latter is controlled by the hand of the pilot, and the buoyancy
of the stationary plate 4
may be neglected in these considerations because said two influences approximately compensate the downwardly directed power of the receptacle 9.
 Besides the lifting power X,
 also the power of the spring 11 acts upon the body 2, in the same direction as said power X, and, furthermore, also the suction power Z
which is produced in the space 7 by the depression
and acts in the opposite direction. These three powers balance each other at normal flight and determine the profile
 of the steadying surface, as illustrated in Figure 2.

Now, if the wind relatively to the flying machine, suddenly grows stronger owing to a squall or gust, then also the powers X and R increase at the same time.
 Although also the suction power at the slot 10
 becomes greater at the same time, still, the power Z does not immediately vary because the amount of air to be led away
 is increased by the supply of air taking place through the aperture 12 and 15.

Besides, the increased suction power at the slot 12 which tends to increase the power Z is opposed by three other powers. Firstly, the increased lifting power X tends
 to turn the body 2 upwards
which is made possible only by air entering through the apertures 12 and 15. Secondly, the flying machine is, at the commencement of the squall
or gust, at first shoved backward by reason
of the additional resistance,
 in consequence of which the high-lying center of gravity of the body 2 and the poise 3 swing forward by reason of their inertia ; thirdly, this latter cause makes the air present
within the channel 13 move
in the direction to the space 7.
A sudden increase of the lifting power X by an outer wind power causes, thus, the body 2 to swing upwards, whereby the additional buoyancy at the front of the machine
is made ineffective.



Matters are different with the rigid main supporting surface at which the power R
 has been increased. The diagram of forces (Figure 3) shows that the front of
the machine has remained at the point C,
 owing to the annihilation of the addition
to the power X, whereas the main supporting plane ascends from A to B.
The swinging motion in forward direction occurring at the rear part of the machine
 being lifted
entails forward drive, whereby the speed of the machine is increased
and, also a variation in the play of forces at the steadying devicer is brought about.
As the angle of incidence at the main supporting plane has become smaller
by reason of the additional upward motion, suction takes place at the slot 15,
and as also the opposing forces that acted upon the body 2 do no more exist,
 this latter is turned in the direction of the plate 4 whereby the buoyancy
at the front of the machine is increased and this latter rises from the position C
to the position D.


From both successively occurring motions of the flying machine results that this latter
is rising for the height H without a loss of speed.In the practical application of the
arrangement in question a greater measure than the length C-A which is chosen only
by way of example will be given the radius of oscillation ;  that may be easily attained
by only partially annihilating,  by the means stated, the power increased by the squall
 or gust.
Owing to its dimension, the rear rim of the movable body 2 forms for the pilot a readily
visible indicator of the conditions of motions of the flying machine.
The pilot may at his discretion either promote the indications of the rear rim or edge
of the body 2 by appropriately operating the horizontal rudder or oppose them,
 also by appropriately actuating said rudder, for instance at landing.
The flying machines with but slight load per unit of surface are easily overturned
by wind after the landing. The soaring flying machines shown in Figure 4 is so
constructedthat the wind cannot lift its front part off the ground because an aperture
 is provided which is kept closed during the flight by a flap 16, but permits of a large
 supply of air through the channel 17 if it is open, so that the steadying plane is
ensured against being drawn against the plate 4.
...
In testimony whereof I have affixed my signature in presence of two witnesses.
FRIEDRICH WHILHELM BUDIG
Witnesses : FRANGE SURNAM and EMIL VORWERK."

 F. Budig montre le dispositif souple appelé ?Harmonica? décrit dans le brevet.

F. Budig a travaillé sur son idée pendant plus de dix ans : il avait déposé un premier brevet relatif à son dispositif de contrôle automatique en France, le 1 août 1914, 
et il a participé à la compétition de planeurs qui devait avoir lieu à Saint-Andreasberg (Allemagne) du 28 janvier au 6 février 1923 et au concours de Rossiten en mai 1924.
(signalons que Hentzen a également participé avec son Wampyr et son Max). Mais en dépit de l'importance et de la durée de ses recherches, il ne semble pas que les résultats 
aient été à la hauteur des espérances de l'inventeur, car les machines de Friedrich Budig sont rapidement tombées dans l'oubli. 

RÉFÉRENCES

[1] The Budig glider, Flight International, 4 janvier 1923. Article + photos
[2]
Start in den Wind, Erlebte Rhöngeschichte 1911-1926, Peter Riedel, Motorbuch Verlag 1977, pages 70, 125
[3] Flying machine, Friedrich W. BUDIG, Brevet US n° 1,419,447 du 13 juin 1922 (reçu le 16 juin 1921)
http://www.freepatentsonline.com/1419447.html  (fichier PDF téléchargeable :
Budig_US1419447.pdf)
[4] Note sur le deuxième concours de Rossiten, Flight International, 26 juin 1924. Note 

Page créée le 11/02/2010. Dernière mise à jour le 15/11/2012.
Des vieilles toiles aux planeurs modernes © ClaudeL 2003 -