Full Text Archive logoFull Text Archive — Free Classic E-books

A History of Aeronautics by E. Charles Vivian

Part 7 out of 8

Adobe PDF icon
Download A History of Aeronautics pdf
File size: 0.9 MB
What's this? light bulb idea Many people prefer to read off-line or to print out text and read from the real printed page. Others want to carry documents around with them on their mobile phones and read while they are on the move. We have created .pdf files of all out documents to accommodate all these groups of people. We recommend that you download .pdfs onto your mobile phone when it is connected to a WiFi connection for reading off-line.

air-cooled, and a 150 horse-power, also of eight cylinders,
water-cooled, running at a normal rate of 1,600 revolutions per
minute. Another notable example of French construction was the
Panhard and Levassor 100 horse-power eight-cylinder Vee engine,
developing its rated power at 1,500 revolutions per minute, and
having the--for that time--low weight of 4.4 lbs. per

American Vee design has followed the British fairly cclosely;
the Curtiss Company produced originally a 75 horse-power
eight-cylinder Vee type running at 1,200 revolutions per minute,
supplementing this with a 170 horse-power engine running at
1,600 revolutions per minute, and later with a twelve-cylinder
model Vee type, developing 300 horse-power at 1,500 revolutions
per minute, with cylinder bore of 5 inches and stroke of 7
inches. An exceptional type of American design was the Kemp Vee
engine of 80 horse-power in which the cylinders were cooled by a
current of air obtained from a fan at the forward end of the
engine. With cylinders of 4.25 inches bore and 4.75 inches
stroke, the rater power was developed at 1,150 revolutions per
minute, and with the engine complete the weight was only 4.75
lbs. per horse-power.


The very first successful design of internal combustion aero
engine made was that of Charles Manly, who built a five-cylinder
radial engine in 1901 for use with Langley's 'aerodrome,' as the
latter inventor decided to call what has since become known as
the aeroplane. Manly made a number of experiments, and finally
decided on radial design, in which the cylinders are so rayed
round a central crank-pin that the pistons act successively upon
it; by this arrangement a very short and compact engine is
obtained, with a minimum of weight, and a regular crankshaft
rotation and perfect balance of inertia forces.

When Manly designed his radial engine, high speed internal
combustion engines were in their infancy, and the difficulties in
construction can be partly realised when the lack of
manufacturing methods for this high-class engine work, and the
lack of experimental data on the various materials, are taken
into account. During its tests, Manly's engine developed 52.4
brake horsepower at a speed of 950 revolutions per minute, with
the remarkably low weight of only 2.4 lbs. per horsepower; this
latter was increased to 3.6 lbs. when the engine was completed by
the addition of ignition system, radiator, petrol tank, and all
accessories, together with the cooling water for the cylinders.

In Manly's engine, the cylinders were of steel, machined outside
and inside to 1/16 of an inch thickness; on the side of cylinder,
at the top end, the valve chamber was brazed, being machined
from a solid forging, The casing which formed the water-jacket
was of sheet steel, 1/50 of an inch in thickness, and this also
was brazed on the cylinder and to the valve chamber. Automatic
inlet valves were fitted, and the exhaust valves were operated
by a cam which had two points, 180 degrees apart; the cam was
rotated in the opposite direction to the engine at one-quarter
engine speed. Ignition was obtained by using a one-spark coil
and vibrator for all cylinders, with a distributor to select the
right cylinder for each spark--this was before the days of the
high-tension magneto and the almost perfect ignition systems that
makers now employ. The scheme of ignition for this engine was
originated by Manly himself, and he also designed the sparking
plugs fitted in the tops of the cylinders. Through fear of
trouble resulting if the steel pistons worked on the steel
cylinders, cast iron liners were introduced in the latter, 1/16
of an inch thick.

The connecting rods of this engine were of virtually the same
type as is employed on nearly all modern radial engines. The
rod for one cylinder had a bearing along the whole of the crank
pin, and its end enclosed the pin; the other four rods had
bearings upon the end of the first rod, and did not touch the
crank pin. The accompanying diagram shows this construction,
together with the means employed for securing the ends of the
four rods--the collars were placed in position after the rods
had been put on. The bearings of these rods did not receive any
of the rubbing effect due to the rotation of the crank pin, the
rubbing on them being only that of the small angular displacement
of the rods during each revolution; thus there was no difficulty
experienced with the lubrication.

Another early example of the radial type of engine was the
French Anzani, of which type one was fitted to the machine with
which Bleriot first crossed the English Channel--this was of 25
horse-power. The earliest Anzani engines were of the
three-cylinder fan type, one cylinder being vertical, and the
other two placed at an angle of 72 degrees on each side, as the
possibility of over-lubrication of the bottom cylinders was
feared if a regular radial construction were adopted. In order
to overcome the unequal balance of this type, balance weights
were fitted inside the crank case.

The final development of this three-cylinder radial was the 'Y'
type of engine, in which the cylinders were regularly disposed
at 120 degrees apart, the bore was 4.1, stroke 4.7 inches, and
the power developed was 30 brake horse-power at 1,300
revolutions per minute.

Critchley's list of aero engines being constructed in 1910 shows
twelve of the radial type, with powers of between 14 and 100
horse-power, and with from three to ten cylinder--this last is
probably the greatest number of cylinders that can be
successfully arranged in circular form. Of the twelve types of
1910, only two were water-cooled, and it is to be noted that
these two ran at the slowest speeds and had the lowest weight per
horse-power of any.

The Anzani radial was considerably developed special attention
being paid to this type by its makers and by 1914 the Anzani
list comprised seven different sizes of air-cooled radials. Of
these the largest had twenty cylinders, developing 200 brake
horse-power--it was virtually a double radial--and the smallest
was the original 30 horse-power three-cylinder design. A
six-cylinder model was formed by a combination of two groups of
three cylinders each, acting upon a double-throw crankshaft; the
two crank pins were set at 180 degrees to each other, and the
cylinder groups were staggered by an amount equal to the
distance between the centres of the crank pins. Ten-cylinder
radial engines are made with two groups of five cylinders acting
upon two crank pins set at 180 degrees to each other, the largest
Anzani 'ten' developed 125 horsepower at 1,200 revolutions per
minute, the ten cylinders being each 4.5 inches in bore with
stroke of 5.9 inches, and the weight of the engine being 3.7 lbs.
per horse-power. In the 200 horse-power Anzani radial the
cylinders are arranged in four groups of five each, acting on two
crank pins. The bore of the cylinders in this engine is the same
as in the three-cylinder, but the stroke is increased to 5.5
inches. The rated power is developed at 1,300 revolutions per
minute, and the engine complete weighs 3.4 lbs. per horse-power.

With this 200 horse-power Anzani, a petrol consumption of as low
as 0.49 lbs. of fuel per brake horse-power per hour has been
obtained, but the consumption of lubricating oil is
compensatingly high, being up to one-fifth of the fuel used. The
cylinders are set desaxe with the crank shaft, and are of
cast-iron, provided with radiating ribs for air-cooling; they are
attached to the crank case by long bolts passing through bosses
at the top of the cylinders, and connected to other bolts at
right angles through the crank case. The tops of the cylinders
are formed flat, and seats for the inlet and exhaust valves are
formed on them. The pistons are cast-iron, fitted with ordinary
cast-iron spring rings. An aluminium crank case is used, being
made in two halves connected together by bolts, which latter also
attach the engine to the frame of the machine. The crankshaft
is of nickel steel, made hollow, and mounted on ball-bearings in
such a manner that practically a combination of ball and plain
bearings is obtained; the central web of the shaft is bent to
bring the centres of the crank pins as close together as
possible, leaving only room for the connecting rods, and the pins
are 180 degrees apart. Nickel steel valves of the cone-seated,
poppet type are fitted, the inlet valves being automatic, and
those for the exhaust cam-operated by means of push-rods. With
an engine having such a number of cylinders a very uniform
rotation of the crankshaft is obtained, and in actual running
there are always five of the cylinders giving impulses to the
crankshaft at the same time.

An interesting type of pioneer radial engine was the Farcot, in
which the cylinders were arranged in a horizontal plane, with a
vertical crankshaft which operated the air-screw through bevel
gearing. This was an eight-cylinder engine, developing 64
horse-power at 1,200 revolutions per minute. The R.E.P. type,in
the early days, was a 'fan' engine, but the designer, M. Robert
Pelterie, turned from this design to a seven-cylinder radial,
which at 1,100 revolutions per minute gave 95 horse-power.
Several makers entered into radial engine development in the
years immediately preceding the War, and in 1914 there were some
twenty-two different sizes and types, ranging from 30 to 600
horse-power, being made, according to report; the actual
construction of the latter size at this time, however, is

Probably the best example of radial construction up to the
outbreak of War was the Salmson (Canton-Unne) water-cooled, of
which in 1914 six sizes were listed as available. Of these
the smallest was a seven-cylinder 90 horse-power engine, and the
largest, rated at 600 horse-power, had eighteen cylinders.
These engines, during the War, were made under license by the
Dudbridge Ironworks in Great Britain.

The accompanying diagram shows the construction of the cylinders
in the 200 horse-power size, showing the method of cooling, and
the arrangement of the connecting rods. A patent planetary gear,
also shown in the diagram, gives exactly the same stroke to all
the pistons. The complete engine has fourteen cylinders, of
forged steel machined all over, and so secured to the crank
case that any one can be removed without parting the crank case.
The water-jackets are of spun copper, brazed on to the cylinder,
and corrugated so as to admit of free expansion; the water is
circulated by means of a centrifugal pump. The pistons are of
cast-iron, each fitted with three rings, and the connecting rods
are of high grade steel, machined all over and fitted with
bushes of phosphor bronze; these rods are connected to a central
collar, carried on the crank pin by two ball-bearings. The
crankshaft has a single throw, and is made in two parts to allow
the cage for carrying the big end-pins of the connecting rods to
be placed in position.

The casing is in two parts, on one of which the brackets for
fixing the engine are carried, while the other part carries the
valve-gear. Bolts secure the two parts together. The
mechanically-operated steel valves on the cylinders are each
fitted with double springs and the valves are operated by rods
and levers. Two Zenith carburettors are fitted on the rear half
of the crank case, and short induction pipes are led to each
cylinder; each of the carburettors is heated by the exhaust
gases. Ignition is by two high-tension magnetos, and a
compressed air self-starting arrangement is provided. Two oil
pumps are fitted for lubricating purposes, one of which forces
oil to the crankshaft and connecting-rod bearings, while the
second forces oil to the valve gear, the cylinders being so
arranged that the oil which flows along the walls cannot flood
the lower cylinders. This engine operates upon a six-stroke
cycle, a rather rare arrangement for internal combustion engines
of the electrical ignition type; this is done in order to obtain
equal angular intervals for the working impulses imparted to the
rotating crankshaft, as the cylinders are arranged in groups of
seven, and all act upon the one crankshaft. The angle,
therefore, between the impulses is 77 1/7 degrees. A diagram is
inset giving a side view of the engine, in order to show the
grouping of the cylinders.

The 600 horse-power Salmson engine was designed with a view to
fitting to airships, and was in reality two nine-cylindered
engines, with a gear-box connecting them; double air-screws were
fitted, and these were so arranged that either or both of them
might be driven by either or both engines; in addition to this,
the two engines were complete and separate engines as regards
carburation and ignition, etc., so that they could be run
independently of each other. The cylinders were exceptionally
'long stroke,' being 5.9 inches bore to 8.27 inches stroke, and
the rated power was developed at 1,200 revolutions per minute,
the weight of the complete engine being only 4.1 lbs. per
horse-power at the normal rating.

A type of engine specially devised for airship propulsion is
that in which the cylinders are arranged horizontally instead of
vertically, the main advantages of this form being the reduction
of head resistance and less obstruction to the view of the
pilot. A casing, mounted on the top of the engine, supports the
air-screw, which is driven through bevel gearing from the upper
end of the crankshaft. With this type of engine a better rate
of air-screw efficiency is obtained by gearing the screw down to
half the rate of revolution of the engine, this giving a more
even torque. The petrol consumption of the type is very low,
being only 0.48 lbs. per horse-power per hour, and equal
economy is claimed as regards lubricating oil, a consumption of
as little as 0.04 lbs. per horse-power per hour being claimed.

Certain American radial engines were made previous to 1914, the
principal being the Albatross six-cylinder engines of 50 and 100
horse-powers. Of these the smaller size was air-cooled, with
cylinders of 4.5 inches bore and 5 inches stroke, developing the
rated power at 1,230 revolutions per minute, with a weight of
about 5 lbs. per horse-power. The 100 horse-power size had
cylinders of 5.5 inches bore, developing its rated power at 1,230
revolutions per minute, and weighing only 2.75 lbs. per
horse-power. This engine was markedly similar to the
six-cylindered Anzani, having all the valves mechanically
operated, and with auxiliary exhaust ports at the bottoms of the
cylinders, overrun by long pistons. These Albatross engines had
their cylinders arranged in two groups of three, with each group
of three pistons operating on one of two crank pins, each
180 degrees apart.

The radial type of engine, thanks to Charles Manly, had the
honour of being first in the field as regards aero work. Its
many advantages, among which may be specially noted the very
short crankshaft as compared with vertical, Vee, or 'broad arrow'
type of engine, and consequent greater rigidity, ensure it
consideration by designers of to-day, and render it certain that
the type will endure. Enthusiasts claim that the 'broad arrow'
type, or Vee with a third row of cylinders inset between the
original two, is just as much a development from the radial
engine as from the vertical and resulting Vee; however this may
be, there is a place for the radial type in air-work for as long
as the internal combustion engine remains as a power plant.


M. Laurent Seguin, the inventor of the Gnome rotary aero engine,
provided as great a stimulus to aviation as any that was given
anterior to the war period, and brought about a great advance in
mechanical flight, since these well-made engines gave a
high-power output for their weight, and were extremely smooth
in running. In the rotary design the crankshaft of the engine
is stationary, and the cylinders, crank case, and all their
adherent parts rotate; the working is thus exactly opposite in
principle to that of the radial type of aero engine, and the
advantage of the rotary lies in the considerable flywheel effect
produced by the revolving cylinders, with consequent evenness of
torque. Another advantage is that air-cooling, adopted in all
the Gnome engines, is rendered much more effective by the
rotation of the cylinders, though there is a tendency to
distortion through the leading side of each cylinder being more
efficiently cooled than the opposite side; advocates of other
types are prone to claim that the air resistance to the
revolving cylinders absorbs some 10 per cent of the power
developed by the rotary engine, but that has not prevented the
rotary from attaining to great popularity as a prime mover.

There were, in the list of aero engines compiled in 1910,
five rotary engines included, all air-cooled. Three of these
were Gnome engines, and two of the make known as 'International.'
They ranged from 21.5 to 123 horse-power, the latter being rated
at only 1.8 lbs. weight per brake horse-power, and having
fourteen cylinders, 4.33 inches in diameter by 4.7 inches stroke.
By 1914 forty-three different sizes and types of rotary engine
were being constructed, and in 1913 five rotary type engines were
entered for the series of aeroplane engine trials held in
Germany. Minor defects ruled out four of these, and only the
German Bayerischer Motoren Flugzeugwerke completed the seven-hour
test prescribed for competing engines. Its large fuel
consumption barred this engine from the final trials, the
consumption being some 0.95 pints per horse-power per hour. The
consumption of lubricating oil, also was excessive, standing at
0.123 pint per horse-power per hour. The engine gave 37.5
effective horse-power during its trial, and the loss due to air
resistance was 4.6 horse-power, about 11 per cent. The
accompanying drawing shows the construction of the engine, in
which the seven cylinders are arranged radially on the crank
case; the method of connecting the pistons to the crank pins can
be seen. The mixture is drawn through the crank chamber, and to
enter the cylinder it passes through the two automatic valves in
the crown of the piston; the exhaust valves are situated in the
tops of the cylinders, and are actuated by cams and push-rods.
Cooling of the cylinder is assisted by the radial rings, and the
diameter of these rings is increased round the hottest part of
the cylinder. When long flights are undertaken the advantage of
the light weight of this engine is more than counterbalanced by
its high fuel and lubricating oil consumption, but there are
other makes which are much better than this seven-cylinder German
in respect of this.

Rotation of the cylinders in engines of this type is produced by
the side pressure of the pistons on the cylinder walls, and in
order to prevent this pressure from becoming abnormally large it
is necessary to keep the weight of the piston as low as possible,
as the pressure is produced by the tangential acceleration and
retardation of the piston. On the upward stroke the
circumferential velocity of the piston is rapidly increased,
which causes it to exert a considerable tangential pressure on
the side of the cylinder, and on the return stroke there is a
corresponding retarding effect due to the reduction of the
circumferential velocity of the piston. These side pressures
cause an appreciable increase in the temperatures of the
cylinders and pistons, which makes it necessary to keep the
power rating of the engines fairly low.

Seguin designed his first Gnome rotary as a 34 horse-power
engine when run at a speed of 1,300 revolutions per minute. It
had five cylinders, and the weight was 3.9 lbs. per horse-power.
A seven-cylinder model soon displaced this first engine, and
this latter, with a total weight of 165 lbs., gave 61.5
horse-power. The cylinders were machined out of solid nickel
chrome-steel ingots, and the machining was carried out so that
the cylinder walls were under 1/6 of an inch in thickness. The
pistons were cast-iron, fitted each with two rings, and the
automatic inlet valve to the cylinder was placed in the crown of
the piston. The connecting rods, of 'H' section, were of nickel
chrome-steel, and the large end of one rod, known as the
'master-rod' embraced the crank pin; on the end of this rod six
hollow steel pins were carried, and to these the remaining six
connecting-rods were attached. The crankshaft of the engine was
made of nickel chrome-steel, and was in two parts connected
together at the crank pin; these two parts, after the master-rod
had been placed in position and the other connecting rods had
been attached to it, were firmly secured. The steel crank case
was made in five parts, the two central ones holding the
cylinders in place, and on one side another of the five castings
formed a cam-box, to the outside of which was secured the
extension to which the air-screw was attached. On the other
side of the crank case another casting carried the thrust-box,
and the whole crank case, with its cylinders and gear, was
carried on the fixed crank shaft by means of four ball-bearings,
one of which also took the axial thrust of the air-screw.

For these engines, castor oil is the lubricant usually adopted,
and it is pumped to the crankshaft by means of a gear-driven oil
pump; from this shaft the other parts of the engine are
lubricated by means of centrifugal force, and in actual practice
sufficient unburnt oil passes through the cylinders to lubricate
the exhaust valve, which partly accounts for the high rate of
consumption of lubricating oil. A very simple carburettor of
the float less, single-spray type was used, and the mixture was
passed along the hollow crankshaft to the interior of the crank
case, thence through the automatic inlet valves in the tops of
the pistons to the combustion chambers of the cylinders.
Ignition was by means of a high-tension magneto specially geared
to give the correct timing, and the working impulses occurred at
equal angular intervals of 102.85 degrees. The ignition was
timed so that the firing spark occurred when the cylinder was 26
degrees before the position in which the piston was at the outer
end of its stroke, and this timing gave a maximum pressure in
the cylinder just after the piston had passed this position.

By 1913, eight different sizes of the Gnome engine were being
constructed, ranging from 45 to 180 brake horse-power; four of
these were single-crank engines one having nine and the other
three having seven cylinders. The remaining four were
constructed with two cranks; three of them had fourteen
cylinders apiece, ranged in groups of seven, acting on the
cranks, and the one other had eighteen cylinders ranged in two
groups of nine, acting on its two cranks. Cylinders of the
two-crank engines are so arranged (in the fourteen-cylinder
type) that fourteen equal angular impulses occur during each
cycle; these engines are supported on bearings on both sides of
the engine, the air-screw being placed outside the front
support. In the eighteen-cylinder model the impulses occur at
each 40 degrees of angular rotation of the cylinders, securing
an extremely even rotation of the air-screw.

In 1913 the Gnome Monosoupape engine was introduced, a model in
which the inlet valve to the cylinder was omitted, while the
piston was of the ordinary cast-iron type. A single exhaust
valve in the cylinder head was operated in a manner similar to
that on the previous Gnome engines, and the fact of this being
the only valve on the cylinder gave the engine its name. Each
cylinder contained ports at the bottom which communicated with
the crank chamber, and were overrun by the piston when this
was approaching the bottom end of its stroke. During the
working cycle of the engine the exhaust valve was opened early
to allow the exhaust gases to escape from the cylinder, so that
by the time the piston overran the ports at the bottom the
pressure within the cylinder was approximately equal to that in
the crank case, and practically no flow of gas took place in
either direction through the ports. The exhaust valve remained
open as usual during the succeeding up-stroke of the piston, and
the valve was held open until the piston had returned through
about one-third of its downward stroke, thus permitting fresh air
to enter the cylinder. The exhaust valve then closed, and the
downward motion of the piston, continuing, caused a partial
vacuum inside the cylinder; when the piston overran the ports,
the rich mixture from the crank case immediately entered. The
cylinder was then full of the mixture, and the next upward stroke
of the piston compressed the charge; upon ignition the working
cycle was repeated. The speed variation of this engine was
obtained by varying the extent and duration of the opening of the
exhaust valves, and was controlled by the pilot by hand-operated
levers acting on the valve tappet rollers. The weight per
horsepower of these engines was slightly less than that of the
two-valve type, while the lubrication of the gudgeon pin and
piston showed an improvement, so that a lower lubricating oil
consumption was obtained. The 100 horse-power Gnome Monosoupape
was built with nine cylinders, each 4.33 inches bore by 5.9
inches stroke, and it developed its rated power at 1,200
revolutions per minute.

An engine of the rotary type, almost as well known as the Gnome,
is the Clerget, in which both cylinders and crank case are made
of steel, the former having the usual radial fins for cooling.
In this type the inlet and exhaust valves are both located in
the cylinder head, and mechanically operated by push-rods and
rockers. Pipes are carried from the crank case to the inlet
valve casings to convey the mixture to the cylinders, a
carburettor of the central needle type being used. The
carburetted mixture is taken into the crank case chamber in a
manner similar to that of the Gnome engine. Pistons of
aluminium alloy, with three cast-iron rings, are fitted, the top
ring being of the obturator type. The large end of one of the
nine connecting rods embraces the crank pin and the pressure is
taken on two ball-bearings housed in the end of the rod. This
carries eight pins, to which the other rods are attached, and the
main rod being rigid between the crank pin and piston pin
determines the position of the pistons. Hollow connecting-rods
are used, and the lubricating oil for the piston pins passes from
the crankshaft through the centres of the rods. Inlet and
exhaust valves can be set quite independently of one another--a
useful point, since the correct timing of the opening of these
valves is of importance. The inlet valve opens 4 degrees from
top centre and closes after the bottom dead centre of the piston;
the exhaust valve opens 68 degrees before the bottom centre and
closes 4 degrees after the top dead centre of the piston. The
magnetos are set to give the spark in the cylinder at 25 degrees
before the end of the compression stroke--two high-tension
magnetos are used: if desired, the second one can be adjusted to
give a later spark for assisting the starting of the engine. The
lubricating oil pump is of the valveless two-plunger type, so
geared that it runs at seven revolutions to 100 revolutions of
the engine; by counting the pulsations the speed of the engine
can be quickly calculated by multiplying the pulsations by 100
and dividing by seven. In the 115 horse-power nine-cylinder
Clerget the cylinders are 4.7 bore with a 6.3 inches stroke, and
the rated power of the engine is obtained at 1,200 revolutions
per minute. The petrol consumption is 0.75 pint per horse-power
per hour.

A third rotary aero engine, equally well known with the
foregoing two, is the Le Rhone, made in four different sizes
with power outputs of from 50 to 160 horse-power; the two
smaller sizes are single crank engines with seven and nine
cylinders respectively, and the larger sizes are of double-crank
design, being merely the two smaller sizes doubled--fourteen and
eighteen-cylinder engines. The inlet and exhaust valves are
located in the cylinder head, and both valves are mechanically
operated by one push-rod and rocker, radial pipes from crank
case to inlet valve casing taking the mixture to the cylinders.
The exhaust valves are placed on the leading, or air-screw side,
of the engine, in order to get the fullest possible cooling
effect. The rated power of each type of engine is obtained at
1,200 revolutions per minute, and for all four sizes the
cylinder bore is 4.13 inches, with a 5.5 inches piston stroke.
Thin cast-iron liners are shrunk into the steel cylinders in
order to reduce the amount of piston friction. Although the Le
Rhone engines are constructed practically throughout of steel,
the weight is only 2.9 lbs. per horse-power in the
eighteen-cylinder type.

American enterprise in the construction of the rotary type is
perhaps best illustrated in the 'Gyro 'engine; this was first
constructed with inlet valves in the heads of the pistons, after
the Gnome pattern, the exhaust valves being in the heads of the
cylinders. The inlet valve in the crown of each piston was
mechanically operated in a very ingenious manner by the
oscillation of the connecting-rod. The Gyro-Duplex engine
superseded this original design, and a small cross-section
illustration of this is appended. It is constructed in seven and
nine-cylinder sizes, with a power range of from 50 to 100
horse-power; with the largest size the low weight of 2.5 lbs..
per horse-power is reached. The design is of considerable
interest to the internal combustion engineer, for it embodies a
piston valve for controlling auxiliary exhaust ports, which also
acts as the inlet valve to the cylinder. The piston uncovers the
auxiliary ports when it reaches the bottom of its stroke, and at
the end of the power stroke the piston is in such a position that
the exhaust can escape over the top of it. The exhaust valve in
the cylinder head is then opened by means of the push-rod and
rocker, and is held open until the piston has completed its
upward stroke and returned through more than half its subsequent
return stroke. When the exhaust valve closes, the cylinder has a
charge of fresh air, drawn in through the exhaust valve, and the
further motion of the piston causes a partial vacuum; by the time
the piston reaches bottom dead centre the piston-valve has moved
up to give communication between the cylinder and the crank case,
therefore the mixture is drawn into the cylinder. Both the
piston valve and exhaust valve are operated by cams formed on the
one casting, which rotates at seven-eighths engine speed for the
seven-cylinder type, and nine-tenths engine speed for the
nine-cylinder engines. Each of these cams has four or five
points respectively, to suit the number of cylinders.

The steel cylinders are machined from solid forgings and
provided with webs for air-cooling as shown. Cast-iron pistons
are used, and are connected to the crankshaft in the same manner
as with the Gnome and Le Rhone engines. Petrol is sprayed into
the crank case by a small geared pump and the mixture is taken
from there to the piston valves by radial pipes. Two separate
pumps are used for lubrication, one forcing oil to the crank-pin
bearing and the other spraying the cylinders.

Among other designs of rotary aero engines the E.J.C. is
noteworthy, in that the cylinders and crank case of this engine
rotate in opposite directions, and two air-screws are used, one
being attached to the end of the crankshaft, and the other to the
crank case. Another interesting type is the Burlat rotary, in
which both the cylinders and crankshaft rotate in the same
direction, the rotation of the crankshaft being twice that of the
cylinders as regards speed. This engine is arranged to work on
the four-stroke cycle with the crankshaft making four, and the
cylinders two, revolutions per cycle.

It would appear that the rotary type of engine is capable of but
little more improvement--save for such devices as these of the
last two engines mentioned, there is little that Laurent Seguin
has not already done in the Gnome type. The limitation of the
rotary lies in its high fuel and lubricating oil consumption,
which renders it unsuited for long-distance aero work; it was,
in the war period, an admirable engine for such short runs as
might be involved in patrol work 'over the lines,' and for
similar purposes, but the watercooled Vee or even vertical, with
its much lower fuel consumption, was and is to be preferred for
distance work. The rotary air-cooled type has its uses, and for
them it will probably remain among the range of current types
for some time to come. Experience of matters aeronautical is
sufficient to show, however, that prophecy in any direction is
most unsafe.


Among the first internal combustion engines to be taken into use
with aircraft were those of the horizontally-opposed four-stroke
cycle type, and, in every case in which these engines were used,
their excellent balance and extremely even torque rendered them
ideal-until the tremendous increase in power requirements
rendered the type too long and bulky for placing in the fuselage
of an aeroplane. As power increased, there came a tendency
toward placing cylinders radially round a central crankshaft,
and, as in the case of the early Anzani, it may be said that the
radial engine grew out of the horizontal opposed piston type.
There were, in 1910--that is, in the early days of small power
units, ten different sizes of the horizontally opposed engine
listed for manufacture, but increase in power requirements
practically ruled out the type for air work.

The Darracq firm were the leading makers of these engines in
1910; their smallest size was a 24 horsepower engine, with two
cylinders each of 5.1 inches bore by 4.7 inches stroke. This
engine developed its rated power at 1,500 revolutions per
minute, and worked out at a weight of 5 lbs. per horse-power.
With these engines the cranks are so placed that two regular
impulses are given to the crankshaft for each cycle of working,
an arrangement which permits of very even balancing of the
inertia forces of the engine. The Darracq firm also made a
four-cylindered horizontal opposed piston engine, in which two
revolutions were given to the crankshaft per revolution, at
equal angular intervals.

The Dutheil-Chambers was another engine of this type, and had
the distinction of being the second largest constructed. At
1,000 revolutions per minute it developed 97 horse-power; its
four cylinders were each of 4.93 inches bore by 11.8 inches
stroke--an abnormally long stroke in comparison with the bore.
The weight--which owing to the build of the engine and its length
of stroke was bound to be rather high, actually amounted to 8.2
lbs. per horse-power. Water cooling was adopted, and the engine
was, like the Darracq four-cylinder type, so arranged as to give
two impulses per revolution at equal angular intervals of
crankshaft rotation.

One of the first engines of this type to be constructed in
England was the Alvaston, a water-cooled model which was made in
20, 30, and 50 brake horse-power sizes, the largest being a
four-cylinder engine. All three sizes were constructed to run
at 1,200 revolutions per minute. In this make the cylinders
were secured to the crank case by means of four long tie bolts
passing through bridge pieces arranged across the cylinder
heads, thus relieving the cylinder walls of all longitudinal
explosion stresses. These bridge pieces were formed from chrome
vanadium steel and milled to an 'H' section, and the bearings
for the valve-tappet were forged solid with them. Special
attention was given to the machining of the interiors of the
cylinders and the combustion heads, with the result that the
exceptionally high compression of 95 lbs. per square inch was
obtained, giving a very flexible engine. The cylinder heads
were completely water-jacketed, and copper water-jackets were
also fitted round the cylinders. The mechanically operated
valves were actuated by specially shaped cams, and were so
arranged that only two cams were required for the set of eight
valves. The inlet valves at both ends of the engine were
connected by a single feed-pipe to which the carburettor was
attached, the induction piping being arranged above the engine
in an easily accessible position. Auxiliary air ports were
provided in the cylinder walls so that the pistons overran them
at the end of their stroke. A single vertical shaft running in
ball-bearings operated the valves and water circulating pump,
being driven by spiral gearing from the crankshaft at half
speed. In addition to the excellent balance obtained with this
engine, the makers claimed with justice that the number of
working parts was reduced to an absolute minimum.

In the two-cylinder Darracq, the steel cylinders were machined
from solid, and auxiliary exhaust ports, overrun by the piston
at the inner end of its stroke, were provided in the cylinder
walls, consisting of a circular row of drilled holes--this
arrangement was subsequently adopted on some of the Darracq
racing car engines. The water jackets were of copper, soldered
to the cylinder walls; both the inlet and exhaust valves were
located in the cylinder heads, being operated by rockers and
push-rods actuated by cams on the halftime shaft driven from one
end of the crankshaft. Ignition was by means of a high-tension
magneto, and long induction pipes connected the-ends of the
cylinders to the carburettor, the latter being placed underneath
the engine. Lubrication was effected by spraying oil into the
crank case by means of a pump, and a second pump circulated the
cooling water.

Another good example of this type of engine was the Eole, which
had eight opposed pistons, each pair of which was actuated by a
common combustion chamber at the centre of the engine, two
crankshafts being placed at the outer ends of the engine. This
reversal of the ordinary arrangement had two advantages; it
simplified induction, and further obviated the need for cylinder
heads, since the explosion drove at two piston heads instead of
at one piston head and the top of the cylinder; against this,
however, the engine had to be constructed strongly enough to
withstand the longitudinal stresses due to the explosions, as
the cranks are placed on the outer ends and the cylinders and
crank-cases take the full force of each explosion. Each
crankshaft drove a separate air-screw.

This pattern of engine was taken up by the Dutheil-Chambers firm
in the pioneer days of aircraft, when the firm in question
produced seven different sizes of horizontal engines. The
Demoiselle monoplane used by Santos-Dumont in 1909 was fitted
with a two-cylinder, horizontally-opposed Dutheil-Chambers
engine, which developed 25 brake horse-power at a speed of
1,100 revolutions per minute, the cylinders being of 5 inches
bore by 5.1 inches stroke, and the total weight of the engine
being some 120 lbs. The crankshafts of these engines were
usually fitted with steel flywheels in order to give a very even
torque, the wheels being specially constructed with wire spokes.
In all the Dutheil-Chambers engines water cooling was adopted,
and the cylinders were attached to the crank cases by means of
long bolts passing through the combustion heads.

For their earliest machines, the Clement-Bayard firm constructed
horizontal engines of the opposed piston type. The best known of
these was the 30 horse-power size, which had cylinders of 4.7
inches diameter by 5.1 inches stroke, and gave its rated power
at 1,200 revolutions per minute. In this engine the steel
cylinders were secured to the crank case by flanges, and
radiating ribs were formed around the barrel to assist the
air-cooling. Inlet and exhaust valves were actuated by
push-rods and rockers actuated from the second motion shaft
mounted above the crank case; this shaft also drove the
high-tension magneto with which the engine was fitted. A ring
of holes drilled round each cylinder constituted auxiliary ports
which the piston uncovered at the inner end of its stroke, and
these were of considerable assistance not only in expelling
exhaust gases, but also in moderating the temperature of the
cylinder and of the main exhaust valve fitted in the cylinder
head. A water-cooled Clement-Bayard horizontal engine was also
made, and in this the auxiliary exhaust ports were not embodied;
except in this particular, the engine was very similar to the
water-cooled Darracq.

The American Ashmusen horizontal engine, developing 100
horse-power, is probably the largest example of this type
constructed. It was made with six cylinders arranged on each
side of a common crank case, with long bolts passing through the
cylinder heads to assist in holding them down. The induction
piping and valve-operating gear were arranged below the engine,
and the half-speed shaft carried the air-screw.

Messrs Palons and Beuse, Germans, constructed a light-weight,
air-cooled, horizontally-opposed engine, two-cylindered. In
this the cast-iron cylinders were made very thin, and were
secured to the crank case by bolts passing through lugs cast on
the outer ends of the cylinders; the crankshaft was made hollow,
and holes were drilled through the webs of the connecting-rods
in order to reduce the weight. The valves were fitted to the
cylinder heads, the inlet valves being of the automatic type,
while the exhaust valves were mechanically operated from the
cam-shaft by means of rockers and push-rods. Two carburettors
were fitted, to reduce the induction piping to a minimum; one
was attached to each combustion chamber, and ignition was by the
normal high-tension magneto driven from the halftime shaft.

There was also a Nieuport two-cylinder air-cooled horizontal
engine, developing 35 horse-power when running at 1,300
revolutions per minute, and being built at a weight of 5.1 lbs.
per horse-power. The cylinders were of 5.3 inches diameter by
5.9 inches stroke; the engine followed the lines of the Darracq
and Dutheil-Chambers pretty closely, and thus calls for no
special description.

The French Kolb-Danvin engine of the horizontal type, first
constructed in 1905, was probably the first two-stroke cycle
engine designed to be applied to the propulsion of aircraft; it
never got beyond the experimental stage, although its trials
gave very good results. Stepped pistons were adopted, and the
charging pump at one end was used to scavenge the power cylinder
at the other ends of the engine, the transfer ports being formed
in the main casting. The openings of these ports were
controlled at both ends by the pistons, and the location of the
ports appears to have made it necessary to take the exhaust from
the bottom of one cylinder and from the top of the other. The
carburetted mixture was drawn into the scavenging cylinders, and
the usual deflectors were cast on the piston heads to assist in
the scavenging and to prevent the fresh gas from passing out of
the exhaust ports.


Although it has been little used for aircraft propulsion, the
possibilities of the two-stroke cycle engine render some study
of it desirable in this brief review of the various types of
internal combustion engine applicable both to aeroplanes and
airships. Theoretically the two-stroke cycle engine--or as it
is more commonly termed, the 'two-stroke,' is the ideal power
producer; the doubling of impulses per revolution of the
crankshaft should render it of very much more even torque than
the four-stroke cycle types, while, theoretically, there should
be a considerable saving of fuel, owing to the doubling of the
number of power strokes per total of piston strokes. In
practice, however, the inefficient scavenging of virtually every
two-stroke cycle engine produced nullifies or more than
nullifies its advantages over the four-stroke cycle engine; in
many types, too, there is a waste of fuel gases through the
exhaust ports, and much has yet to be done in the way of
experiment and resulting design before the two-stroke cycle
engine can be regarded as equally reliable, economical, and
powerful with its elder brother.

The first commercially successful engine operating on the
two-stroke cycle was invented by Mr Dugald Clerk, who in 1881
proved the design feasible. As is more or less generally
understood, the exhaust gases of this engine are discharged from
the cylinder during the time that the piston is passing the
inner dead centre, and the compression, combustion, and
expansion of the charge take place in similar manner to that of
the four-stroke cycle engine. The exhaust period is usually
controlled by the piston overrunning ports in the cylinder at
the end of its working stroke, these ports communicating direct
with the outer air--the complication of an exhaust valve is thus
obviated; immediately after the escape of the exhaust gases,
charging of the cylinder occurs, and the fresh gas may be
introduced either through a valve in the cylinder head or
through ports situated diametrically opposite to the exhaust
ports. The continuation of the outward stroke of the piston,
after the exhaust ports have been closed, compresses the charge
into the combustion chamber of the cylinder, and the ignition of
the mixture produces a recurrence of the working stroke.

Thus, theoretically, is obtained the maximum of energy with the
minimum of expenditure; in practice, however, the scavenging of
the power cylinder, a matter of great importance in all internal
combustion engines, is often imperfect, owing to the opening of
the exhaust ports being of relatively short duration; clearing
the exhaust gases out of the cylinder is not fully accomplished,
and these gases mix with the fresh charge and detract from its
efficiency. Similarly, owing to the shorter space of time
allowed, the charging of the cylinder with the fresh mixture is
not so efficient as in the four-stroke cycle type; the fresh
charge is usually compressed slightly in a separate
chamber--crank case, independent cylinder, or charging pump, and
is delivered to the working cylinder during the beginning of the
return stroke of the piston, while in engines working on the
four-stroke cycle principle a complete stroke is devoted to the
expulsion of the waste gases of the exhaust, and another full
stroke to recharging the cylinder with fresh explosive mixture.

Theoretically the two-stroke and the four-stroke cycle engines
possess exactly the same thermal efficiency, but actually this
is modified by a series of practical conditions which to some
extent tend to neutralise the very strong case in favour of the
two-stroke cycle engine. The specific capacity of the engine
operating on the two-stroke principle is theoretically twice
that of one operating on the four-stroke cycle, and
consequently, for equal power, the former should require only
about half the cylinder volume of the latter; and, owing to the
greater superficial area of the smaller cylinder, relatively,
the latter should be far more easily cooled than the larger
four-stroke cycle cylinder; thus it should be possible to get
higher compression pressures, which in turn should result in
great economy of working. Also the obtaining of a working
impulse in the cylinder for each revolution of the crankshaft
should give a great advantage in regularity of rotation--which
it undoubtedly does--and the elimination of the operating gear
for the valves, inlet and exhaust, should give greater
simplicity of design.

In spite of all these theoretical--and some practical--advantages
the four-stroke cycle engine was universally adopted for aircraft
work; owing to the practical equality of the two principles of
operation, so far as thermal efficiency and friction losses are
concerned, there is no doubt that the simplicity of design (in
theory) and high power output to weight ratio (also in theory)
ought to have given the 'two-stroke' a place on the aeroplane.
But this engine has to be developed so as to overcome its
inherent drawbacks; better scavenging methods have yet to be
devised--for this is the principal drawback--before the
two-stroke can come to its own as a prime mover for aircraft.

Mr Dugald Clerk's original two-stroke cycle engine is indicated
roughly, as regards principle, by the accompanying diagram, from
which it will be seen that the elimination of the ordinary inlet
and exhaust valves of the four-stroke type is more than
compensated by a separate cylinder which, having a piston worked
from the connecting-rod of the power cylinder, was used to
charging, drawing the mixture from the carburettor past the
valve in the top of the charging cylinder, and then forcing it
through the connecting pipe into the power cylinder. The inlet
valves both on the charging and the power cylinders are
automatic; when the power piston is near the bottom of its
stroke the piston in the charging cylinder is compressing the
carburetted air, so that as soon as the pressure within the
power cylinder is relieved by the exit of the burnt gases
through the exhaust ports the pressure in the charging cylinder
causes the valve in the head of the power cylinder to open, and
fresh mixture flows into the cylinder, replacing the exhaust
gases. After the piston has again covered the exhaust ports the
mixture begins to be compressed, thus automatically closing the
inlet valve. Ignition occurs near the end of the compression
stroke, and the working stroke immediately follows, thus giving
an impulse to the crankshaft on every down stroke of the piston.
If the scavenging of the cylinder were complete, and the cylinder
were to receive a full charge of fresh mixture for every stroke,
the same mean effective pressure as is obtained with four-stroke
cycle engines ought to be realised, and at an equal speed of
rotation this engine should give twice the power obtainable from
a four-stroke cycle engine of equal dimensions. This result was
not achieved, and, with the improvements in construction brought
about by experiment up to 1912, the output was found to be only
about fifty per cent more than that of a four-stroke cycle engine
of the same size, so that, when the charging cylinder is
included, this engine has a greater weight per horse-power, while
the lowest rate of fuel consumption recorded was 0.68 lb. per
horse-power per hour.

In 1891 Mr Day invented a two-stroke cycle engine which used the
crank case as a scavenging chamber, and a very large number of
these engines have been built for industrial purposes. The
charge of carburetted air is drawn through a non-return valve
into the crank chamber during the upstroke of the piston, and
compressed to about 4 lbs. pressure per square inch on the
down stroke. When the piston approaches the bottom end of its
stroke the upper edge first overruns an exhaust port, and almost
immediately after uncovers an inlet port on the opposite side of
the cylinder and in communication with the crank chamber; the
entering charge, being under pressure, assists in expelling the
exhaust gases from the cylinder. On the next upstroke the
charge is compressed into the combustion space of the cylinder,
a further charge simultaneously entering the crank case to be
compressed after the ignition for the working stroke. To
prevent the incoming charge escaping through the exhaust ports
of the cylinder a deflector is formed on the top of the piston,
causing the fresh gas to travel in an upward direction, thus
avoiding as far as possible escape of the mixture to the
atmosphere. From experiments conducted in 1910 by Professor
Watson and Mr Fleming it was found that the proportion of fresh
gases which escaped unburnt through the exhaust ports diminished
with increase of speed; at 600 revolutions per minute about 36
per cent of the fresh charge was lost; at 1,200 revolutions per
minute this was reduced to 20 per cent, and at 1,500 revolutions
it was still farther reduced to 6 per cent.

So much for the early designs. With regard to engines of this
type specially constructed for use with aircraft, three designs
call for special mention. Messrs A. Gobe and H. Diard, Parisian
engineers, produced an eight-cylindered two-stroke cycle engine
of rotary design, the cylinders being co-axial. Each pair of
opposite pistons was secured together by a rigid connecting rod,
connected to a pin on a rotating crankshaft which was mounted
eccentrically to the axis of rotation of the cylinders. The
crankshaft carried a pinion gearing with an internally toothed
wheel on the transmission shaft which carried the air-screw. The
combustible mixture, emanating from a common supply pipe, was led
through conduits to the front ends of the cylinders, in which the
charges were compressed before being transferred to the working
spaces through ports in tubular extensions carried by the
pistons. These extensions had also exhaust ports, registering
with ports in the cylinder which communicated with the outer air,
and the extensions slid over depending cylinder heads attached to
the crank case by long studs. The pump charge was compressed in
one end of each cylinder, and the pump spaces each delivered
into their corresponding adjacent combustion spaces. The charges
entered the pump spaces during the suction period through
passages which communicated with a central stationary supply
passage at one end of the crank case, communication being cut off
when the inlet orifice to the passage passed out of register with
the port in the stationary member. The exhaust ports at the
outer end of the combustion space opened just before and closed a
little later than the air ports, and the incoming charge assisted
in expelling the exhaust gases in a manner similar to that of the
earlier types of two-stroke cycle engine; The accompanying rough
diagram assists in showing the working of this engine.

Exhibited in the Paris Aero Exhibition of 1912, the Laviator
two-stroke cycle engine, six-cylindered, could be operated either
as a radial or as a rotary engine, all its pistons acting on a
single crank. Cylinder dimensions of this engine were 3.94
inches bore by 5.12 inches stroke, and a power output of 50
horse-power was obtained when working at a rate of 1,200
revolutions per minute. Used as a radial engine, it developed
65 horse-power at the same rate of revolution, and, as the total
weight was about 198 lbs., the weight of about 3 lbs. per
horse-power was attained in radial use. Stepped pistons were
employed, the annular space between the smaller or power piston
and the walls of the larger cylinder being used as a charging
pump for the power cylinder situated 120 degrees in rear of it.
The charging cylinders were connected by short pipes to ports in
the crank case which communicated with the hollow crankshaft
through which the fresh gas was supplied, and once in each
revolution each port in the case registered with the port in the
hollow shaft. The mixture which then entered the charging
cylinder was transferred to the corresponding working
cylinder when the piston of that cylinder had reached the end of
its power stroke, and immediately before this the exhaust ports
diametrically opposite the inlet ports were uncovered; scavenging
was thus assisted in the usual way. The very desirable feature
of being entirely valveless was accomplished with this engine,
which is also noteworthy for exceedingly compact design.

The Lamplough six-cylinder two-stroke cycle rotary, shown at the
Aero Exhibition at Olympia in 1911, had several innovations,
including a charging pump of rotary blower type. With the six
cylinders, six power impulses at regular intervals were given on
each rotation; otherwise, the cycle of operations was carried
out much as in other two-stroke cycle engines. The pump
supplied the mixture under slight pressure to an inlet port in
each cylinder, which was opened at the same time as the exhaust
port, the period of opening being controlled by the piston. The
rotary blower sucked the mixture from the carburettor and
delivered it to a passage communicating with the inlet ports in
the cylinder walls. A mechanically-operated exhaust valve was
placed in the centre of each cylinder head, and towards the end
of the working stroke this valve opened, allowing part of the
burnt gases to escape to the atmosphere; the remainder was
pushed out by the fresh mixture going in through the ports at
the bottom end of the cylinder. In practice, one or other of
the cylinders was always taking fresh mixture while working,
therefore the delivery from the pump was continuous and the
mixture had not to be stored under pressure.

The piston of this engine was long enough to keep the ports
covered when it was at the top of the stroke, and a bottom ring
was provided to prevent the mixture from entering the crank
case. In addition to preventing leakage, this ring no doubt
prevented an excess of oil working up the piston into the
cylinder. As the cylinder fired with every revolution, the
valve gear was of the simplest construction, a fixed cam lifting
each valve as the cylinder came into position. The spring of
the exhaust valve was not placed round the stem in the usual
way, but at the end of a short lever, away from the heat of the
exhaust gases. The cylinders were of cast steel, the crank case
of aluminium, and ball-bearings were fitted to the crankshaft,
crank pins, and the rotary blower pump. Ignition was by means
of a high-tension magneto of the two-spark pattern, and with a
total weight of 300 lbs. the maximum output was 102 brake
horse-power, giving a weight of just under 3 lbs. per

One of the most successful of the two-stroke cycle engines was
that designed by Mr G. F. Mort and constructed by the New
Engine Company. With four cylinders of 3.69 inches bore by 4.5
inches stroke, and running at 1,250 revolutions per minute, this
engine developed 50 brake horse-power; the total weight of the
engine was 155 lbs., thus giving a weight of 3.1 lbs. per
horse-power. A scavenging pump of the rotary type was employed,
driven by means of gearing from the engine crankshaft, and in
order to reduce weight to a minimum the vanes were of aluminium.
This engine was tried on a biplane, and gave very satisfactory

American design yields two apparently successful two-stroke
cycle aero engines. A rotary called the Fredericson engine was
said to give an output of 70 brake horse-power with five
cylinders 4.5 inches diameter by 4.75 inches stroke, running
at 1,000 revolutions per minute. Another, the Roberts
two-stroke cycle engine, yielded 100 brake horse-power from six
cylinders of the stepped piston design; two carburettors, each
supplying three cylinders, were fitted to this engine. Ignition
was by means of the usual high-tension magneto, gear-driven from
the crankshaft, and the engine, which was water-cooled, was of
compact design.

It may thus be seen that the two-stroke cycle type got as far as
actual experiment in air work, and that with considerable
success. So far, however, the greater reliability of the
four-stroke cycle has rendered it practically the only aircraft
engine, and the two-stroke has yet some way to travel before it
becomes a formidable competitor, in spite of its admitted
theoretical and questioned practical advantages.


The principal engines of British, French, and American design
used in the war period and since are briefly described under the
four distinct types of aero engine; such notable examples as the
Rolls-Royce, Sunbeam, and Napier engines have been given special
mention, as they embodied--and still embody--all that is best in
aero engine practice. So far, however, little has been said
about the development of German aero engine design, apart from
the early Daimler and other pioneer makes.

At the outbreak of hostilities in 1914, thanks to subsidies to
contractors and prizes to aircraft pilots, the German aeroplane
industry was in a comparatively flourishing condition. There
were about twenty-two establishments making different types of
heavier-thanair machines, monoplane and biplane, engined for the
most part with the four-cylinder Argus or the six-cylinder
Mercedes vertical type engines, each of these being of 100
horse-power--it was not till war brought increasing demands on
aircraft that the limit of power began to rise. Contemporary
with the Argus and Mercedes were the Austro-Daimler, Benz, and
N.A.G., in vertical design, while as far as rotary types were
concerned there were two, the Oberursel and the Stahlhertz; of
these the former was by far the most promising, and it came to
virtual monopoly of the rotary-engined plane as soon as the war
demand began. It was practically a copy of the famous Gnome
rotary, and thus deserves little description.

Germany, from the outbreak of war, practically, concentrated on
the development of the Mercedes engine; and it is noteworthy
that, with one exception, increase of power corresponding with
the increased demand for power was attained without increasing
the number of cylinders. The various models ranged between 75
and 260 horse-power, the latter being the most recent production
of this type. The exception to the rule was the eight-cylinder
240 horse-power, which was replaced by the 260 horse-power
six-cylinder model, the latter being more reliable and but very
slightly heavier. Of the other engines, the 120 horsepower
Argus and the 160 and 225 horse-power Benz were the most used,
the Oberursel being very largely discarded after the Fokker
monoplane had had its day, and the N.A.G. and Austro-Daimler
Daimler also falling to comparative disuse. It may be said that
the development of the Mercedes engine contributed very largely
to such success as was achieved in the war period by German
aircraft, and, in developing the engine, the builders were
careful to make alterations in such a way as to effect the least
possible change in the design of aeroplane to which they were to
be fitted. Thus the engine base of the 175 horse-power model
coincided precisely with that of the 150 horse-power model, and
the 200 and 240 horse-power models retained the same base
dimensions. It was estimated, in 1918, that well over eighty
per cent of German aircraft was engined with the Mercedes type.

In design and construction, there was nothing abnormal about the
Mercedes engine, the keynote throughout being extreme
reliability and such simplification of design as would permit of
mass production in different factories. Even before the war,
the long list of records set up by this engine formed practical
application of the wisdom of this policy; Bohn's flight of 24
hours 10 minutes, accomplished on July 10th and 11th, 1914,
9is an instance of this--the flight was accomplished on an
Albatross biplane with a 75 horsepower Mercedes engine. The
radial type, instanced in other countries by the Salmson and
Anzani makes, was not developed in Germany; two radial engines
were made in that country before the war, but the Germans seemed
to lose faith in the type under war conditions, or it may have
been that insistence on standardisation ruled out all but the
proved examples of engine.

Details of one of the middle sizes of Mercedes motor, the 176
horse-power type, apply very generally to the whole range; this
size was in use up to and beyond the conclusion of hostilities,
and it may still be regarded as characteristic of modern (1920)
German practice. The engine is of the fixed vertical type,
has six cylinders in line, not off-set, and is water-cooled.
The cam shaft is carried in a special bronze casing, seated on
the immediate top of the cylinders, and a vertical shaft is
interposed between crankshaft and camshaft, the latter being
driven by bevel gearing.

On this vertical connecting-shaft the water pump is located,
serving to steady the motion of the shaft. Extending immediately
below the camshaft is another vertical shaft, driven by bevel
gears from the crank-shaft, and terminating in a worm which
drives the multiple piston oil pumps.

The cylinders are made from steel forgings, as are the valve
chamber elbows, which are machined all over and welded together.
A jacket of light steel is welded over the valve elbows and
attached to a flange on the cylinders, forming a water-cooling
space with a section of about 7/16 of an inch. The cylinder
bore is 5.5 inches, and the stroke 6.29 inches. The cylinders
are attached to the crank case by means of dogs and long through
bolts, which have shoulders near their lower ends and are bolted
to the lower half of the crank chamber. A very light and rigid
structure is thus obtained, and the method of construction won
the flattery of imitation by makers of other nationality.

The cooling system for the cylinders is extremely efficient.
After leaving the water pump, the water enters the top of the
front cylinders and passes successively through each of the six
cylinders of the row; short tubes, welded to the tops of the
cylinders, serve as connecting links in the system. The Panhard
car engines for years were fitted with a similar cooling system,
and the White and Poppe lorry engines were also similarly
fitted; the system gives excellent cooling effect where it is
most needed, round the valve chambers and the cylinder heads.

The pistons are built up from two pieces; a dropped forged steel
piston head, from which depend the piston pin bosses, is
combined with a cast-iron skirt, into which the steel head is
screwed. Four rings are fitted, three at the upper and one at
the lower end of the piston skirt, and two lubricating oil
grooves are cut in the skirt, in addition to the ring grooves.
Two small rivets retain the steel head on the piston skirt after
it has been screwed into position, and it is also welded at two
points. The coefficient of friction between the cast-iron and
steel is considerably less than that which would exist between
two steel parts, and there is less tendency for the skirt to
score the cylinder walls than would be the case if all steel were
used--so noticeable is this that many makers, after giving steel
pistons a trial, discarded them in favour of cast-iron; the Gnome
is an example of this, being originally fitted with a steel
piston carrying a brass ring, discarded in favour of a cast-iron
piston with a percentage of steel in the metal mixture. In the
Le Rhone engine the difficulty is overcome by a cast-iron liner
to the cylinders.

The piston pin of the Mercedes is of chrome nickel steel, and is
retained in the piston by means of a set screw and cotter pin.
The connecting rods, of I section, are very short and rigid,
carrying floating bronze bushes which fit the piston pins at the
small end, and carrying an oil tube on each for conveying oil
from the crank pin to the piston pin.

The crankshaft is of chrome nickel steel, carried on seven
bearings. Holes are drilled through each of the crank pins and
main bearings, for half the diameter of the shaft, and these are
plugged with pressed brass studs. Small holes, drilled through
the crank cheeks, serve to convey lubricant from the main
bearings to the crank pins. The propeller thrust is taken by a
simple ball thrust bearing at the propeller end of the
crankshaft, this thrust bearing being seated in a steel retainer
which is clamped between the two halves of the crank case. At
the forward end of the crankshaft there is mounted a master
bevel gear on six splines; this bevel floats on the splines
against a ball thrust bearing, and, in turn, the thrust is taken
by the crank case cover. A stuffing box prevents the loss of
lubricant out of the front end of the crank chamber, and an oil
thrower ring serves a similar purpose at the propeller end of the
crank chamber.

With a motor speed of 1,450 r.p.m., the vertical shaft at the
forward end of the motor turns at 2,175 r.p.m., this being the
speed of the two magnetos and the water pump. The lower
vertical shaft bevel gear and the magneto driving gear are made
integral with the vertical driving shaft, which is carried in
plain bearings in an aluminium housing. This housing is clamped
to the upper half of the crank case by means of three studs.
The cam-shaft carries eighteen cams, these being the inlet and
exhaust cams, and a set of half compression cams which are
formed with the exhaust cams and are put into action when
required by means of a lever at the forward end of the
cam-shaft. The cam-shaft is hollow, and serves as a channel for
the conveyance of lubricating oil to each of the camshaft
bearings. At the forward end of this shaft there is also
mounted an air pump for maintaining pressure on the fuel supply
tank, and a bevel gear tachometer drive.

Lubrication of the engine is carried out by a full pressure
system. The oil is pumped through a single manifold, with seven
branches to the crankshaft main bearings, and then in turn
through the hollow crankshaft to the connecting-rod big ends and
thence through small tubes, already noted, to the small end
bearings. The oil pump has four pistons and two double valves
driven from a single eccentric shaft on which are mounted four
eccentrics. The pump is continuously submerged in oil; in order
to avoid great variations in pressure in the oil lines there is
a piston operated pressure regulator, cut in between the pump
and the oil lines. The two small pistons of the pump take fresh
oil from a tank located in the fuselage of the machine; one of
these delivers oil to the cam shaft, and one delivers to the
crankshaft; this fresh oil mixes with the used oil, returns to
the base, and back to the main large oil pump cylinders. By
means of these small pump pistons a constant quantity of oil is
kept in the motor, and the oil is continually being freshened by
means of the new oil coming in. All the oil pipes are very
securely fastened to the lower half of the crank case, and some
cooling of the oil is effected by air passing through channels
cast in the crank case on its way to the carburettor.

A light steel manifold serves to connect the exhaust ports of
the cylinders to the main exhaust pipe, which is inclined about
25 degrees from vertical and is arranged to give on to the
atmosphere just over the top of the upper wing of the aeroplane.

As regards carburation, an automatic air valve surrounds the
throat of the carburettor, maintaining normal composition of
mixture. A small jet is fitted for starting and running without
load. The channels cast in the crank chamber, already alluded
to in connection with oil-cooling, serve to warm the air before
it reaches the carburettor, of which the body is water-jacketed.

Ignition of the engine is by means of two Bosch ZH6 magnetos,
driven at a speed of 2,175 revolutions per minute when the engine
is running at its normal speed of 1,450 revolutions. The maximum
advance of spark is 12 mm., or 32 degrees before the top dead
centre, and the firing order of the cylinders is 1,5,3,6,2,4.

The radiator fitted to this engine, together with the
water-jackets, has a capacity of 25 litres of water, it is
rectangular in shape, and is normally tilted at an angle of 30
degrees from vertical. Its weight is 26 kg., and it offers but
slight head resistance in flight.

The radial type of engine, neglected altogether in Germany, was
brought to a very high state of perfection at the end of the
War period by British makers. Two makes, the Cosmos Engineering
Company's 'Jupiter' and 'Lucifer,' and the A.B.C. 'Wasp II' and
'Dragon Fly 1A' require special mention for their light weight
and reliability on trials.

The Cosmos 'Jupiter' was--for it is no longer being made--a 450
horse-power nine-cylinder radial engine, air-cooled, with the
cylinders set in one single row; it was made both geared to
reduce the propeller revolutions relatively to the crankshaft
revolutions, and ungeared; the normal power of the geared type
was 450 horse-power, and the total weight of the engine,
including carburettors, magnetos, etc., was only 757 lbs.; the
engine speed was 1,850 revolutions per minute, and the propeller
revolutions were reduced by the gearing to 1,200. Fitted to a
'Bristol Badger' aeroplane, the total weight was 2,800 lbs.,
including pilot, passenger, two machine-guns, and full military
load; at 7,000 feet the registered speed, with corrections for
density, was 137 miles per hour; in climbing, the first 2,000
feet was accomplished in 1 minute 4 seconds; 4,000 feet was
reached in 2 minutes 10 seconds; 6,000 feet was reached in 3
minutes 33 seconds, and 7,000 feet in 4 minutes 15 seconds.
It was intended to modify the plane design and fit a new
propeller, in order to attain even better results, but, if
trials were made with these modifications, the results are not

The Cosmos 'Lucifer' was a three-cylinder radial type engine of
100 horse-power, inverted Y design, made on the simplest possible
principles with a view to quantity production and extreme
reliability. The rated 100 horse-power was attained at 1,600
revolutions per minute, and the cylinder dimensions were 5.75
bore by 6.25 inches stroke. The cylinders were of aluminium and
steel mixture, with aluminium heads; overhead valves, operated by
push rods on the front side of the cylinders, were fitted, and a
simple reducing gear ran them at half engine speed. The crank
case was a circular aluminium casting, the engine being attached
to the fuselage of the aeroplane by a circular flange situated at
the back of the case; propeller shaft and crankshaft were
integral. Dual ignition was provided, the generator and
distributors being driven off the back end of the engine and the
distributors being easily accessible. Lubrication was by means
of two pumps, one scavenging and one suction, oil being fed under
pressure from the crankshaft. A single carburettor fed all three
cylinders, the branch pipe from the carburettor to the circular
ring being provided with an exhaust heater. The total weight of
the engine, 'all on,' was 280 lbs.

The A.B.C. 'Wasp II,' made by Walton Motors, Limited, is a
seven-cylinder radial, air-cooled engine, the cylinders having a
bore of 4.75 inches and stroke 6.25 inches. The normal brake
horse-power at 1,650 revolutions is 160, and the maximum 200 at
a speed of 1,850 revolutions per minute. Lubrication is by
means of two rotary pumps, one feeding through the hollow
crankshaft to the crank pin, giving centrifugal feed to big end
and thence splash oiling, and one feeding to the nose of the
engine, dropping on to the cams and forming a permanent sump for
the gears on the bottom of the engine nose. Two carburettors
are fitted, and two two-spark magnetos, running at one and
three-quarters engine speed. The total weight of this engine is
350 lbs., or 1.75 lbs. per horse-power. Oil consumption at 1,850
revolutions is .03 pints per horse-power per hour, and petrol
consumption is .56 pints per horsepower per hour. The engine
thus shows as very economical in consumption, as well as very
light in weight.

The A.B.C. 'Dragon Fly 1A 'is a nine-cylinder radial engine
having one overhead inlet and two overhead exhaust valves per
cylinder. The cylinder dimensions are 5.5 inches bore by 6.5
inches stroke, and the normal rate of speed, 1,650 revolutions
per minute, gives 340 horse-power. The oiling is by means of
two pumps, the system being practically identical with that of
the 'Wasp II.' Oil consumption is .021 pints per brake
horse-power per hour, and petrol consumption .56 pints--the
same as that of the 'Wasp II.' The weight of the complete
engine, including propeller boss, is 600 lbs., or 1,765 lbs.
per horse-power.

These A.B.C. radials have proved highly satisfactory on tests,
and their extreme simplicity of design and reliability commend
them as engineering products and at the same time demonstrate
the value, for aero work, of the air-cooled radial
design--when this latter is accompanied by sound workmanship.
These and the Cosmos engines represent the minimum of weight per
horse-power yet attained, together with a practicable degree of
reliability, in radial and probably any aero engine design.



Paris, October 21, 1897.

Report on the trials of M. Clement Ader's aviation apparatus.

M. Ader having notified the Minister of War by letter, July 21,
1897, that the Apparatus of Aviation which he had agreed to
build under the conditions set forth in the convention of July
24th, 1894, was ready, and therefore requesting that trials be
undertaken before a Committee appointed for this purpose as per
the decision of August 4th, the Committee was appointed as

Division General Mensier, Chairman; Division General Delambre,
Inspector General of the Permanent Works of Coast Defence,
Member of the Technical Committee of the Engineering Corps;
Colonel Laussedat, Director of the Conservatoire des Arts et
Metiers; Sarrau, Member of the Institute, Professor of
Mechanical Engineering at the Polytechnic School; Leaute, Member
of the Institute, Professor of Mechanical Engineering at the
Polytechnique School.

Colonel Laussedat gave notice at once that his health and work
as Director of the Conservatoire des Arts et Metiers did not
permit him to be a member of the Committee; the Minister
therefore accepted his resignation on September 24th, and
decided not to replace him.

Later on, however, on the request of the Chairman of the
Committee, the Minister appointed a new member General Grillon,
commanding the Engineer Corps of the Military Government of

To carry on the trials which were to take place at the camp of
Satory, the Minister ordered the Governor of the Military Forces
of Paris to requisition from the Engineer Corps, on the request
of the Chairman of the Committee, the men necessary to prepare
the grounds at Satory.

After an inspection made on the 16th an aerodrome was chosen.
M. Ader's idea was to have it of circular shape with a width of
40 metres and an average diameter of 450 metres. The preliminary
work, laying out the grounds, interior and exterior
circumference, etc., was finished at the end of August; the work
of smoothing off the grounds began September 1st with forty-five
men and two rollers, and was finished on the day of the first
tests, October 12th.

The first meeting of the Committee was held August 18th in M.
Ader's workshop; the object being to demonstrate the machine to
the Committee and give all the information possible on the tests
that were to be held. After a careful examination and after
having heard all the explanations by the inventor which were
deemed useful and necessary, the Committee decided that the
apparatus seemed to be built with a perfect understanding of the
purpose to be fulfilled as far as one could judge from a study
of the apparatus at rest; they therefore authorised M. Ader to
take the machine apart and carry it to the camp at Satory so as
to proceed with the trials.

By letter of August 19th the Chairman made report to the Minister
of the findings of the Committee.

The work on the grounds having taken longer than was anticipated,
the Chairman took advantage of this delay to call the Committee
together for a second meeting, during which M. Ader was to run
the two propulsive screws situated at the forward end of the

The meeting was held October 2nd. It gave the Committee an
opportunity to appreciate the motive power in all its details;
firebox, boiler, engine, under perfect control, absolute
condensation, automatic fuel and feed of the liquid to be
vaporised, automatic lubrication and scavenging; everything, in
a word, seemed well designed and executed.

The weights in comparison with the power of the engine realised
a considerable advance over anything made to date, since the two
engines weighed together realised 42 kg., the firebox and boiler
60 kg., the condenser 15 kg., or a total of 117 kg. for
approximately 40 horse-power or a little less than 3 kg. per

One of the members summed up the general opinion by saying:
'Whatever may be the result from an aviation point of view, a
result which could not be foreseen for the moment, it was
nevertheless proven that from a mechanical point of view M.
Ader's apparatus was of the greatest interest and real
ingeniosity. He expressed a hope that in any case the machine
would not be lost to science.'

The second experiment in the workshop was made in the presence
of the Chairman, the purpose being to demonstrate that the
wings, having a spread of 17 metres, were sufficiently strong
to support the weight of the apparatus. With this object in
view, 14 sliding supports were placed under each one of these,
representing imperfectly the manner in which the wings would
support the machine in the air; by gradually raising the
supports with the slides, the wheels on which the machine rested
were lifted from the ground. It was evident at that time that
the members composing the skeleton of the wings supported the
apparatus, and it was quite evident that when the wings were
supported by the air on every point of their surface, the stress
would be better equalised than when resting on a few supports,
and therefore the resistance to breakage would be considerably

After this last test, the work on the ground being practically
finished, the machine was transported to Satory, assembled and
again made ready for trial.

At first M. Ader was to manoeuvre the machine on the ground at
a moderate speed, then increase this until it was possible to
judge whether there was a tendency for the machine to rise; and
it was only after M. Ader had acquired sufficient practice that
a meeting of the Committee was to be called to be present at the
first part of the trials; namely, volutions of the apparatus on
the ground.

The first test took place on Tuesday, October 12th, in the
presence of the Chairman of the Committee. It had rained a good
deal during the night and the clay track would have offered
considerable resistance to the rolling of the machine;
furthermore, a moderate wind was blowing from the south-west,
too strong during the early part of the afternoon to allow of
any trials.

Toward sunset, however, the wind having weakened, M. Ader
decided to make his first trial; the machine was taken out of
its hangar, the wings were mounted and steam raised. M. Ader
in his seat had, on each side of him, one man to the right and
one to the left, whose duty was to rectify the direction of the
apparatus in the event that the action of the rear wheel as a
rudder would not be sufficient to hold the machine in a straight

At 5.25 p.m. the machine was started, at first slowly and then
at an increased speed; after 250 or 300 metres, the two men who
were being dragged by the apparatus were exhausted and forced to
fall flat on the ground in order to allow the wings to pass over
them, and the trip around the track was completed, a total of
1,400 metres, without incident, at a fair speed, which could be
estimated to be from 300 to 400 metres per minute.
Notwithstanding M. Ader's inexperience, this being the first
time that he had run his apparatus, he followed approximately
the chalk line which marked the centre of the track and he
stopped at the exact point from which he started.

The marks of the wheels on the ground, which was rather soft,
did not show up very much, and it was clear that a part of the
weight of the apparatus had been supported by the wings, though
the speed was only about one-third of what the machine could do
had M. Ader used all its motive power; he was running at a
pressure of from 3 to 4 atmospheres, when he could have used 10
to 12.

This first trial, so fortunately accomplished, was of great
importance; it was the first time that a comparatively heavy
vehicle (nearly 400 kg., including the weight of the operator,
fuel, and water) had been set in motion by a tractive apparatus,
using the air solely as a propelling medium. The favourable
report turned in by the Committee after the meeting of October
2nd was found justified by the results demonstrated on the
grounds, and the first problem of aviation, namely, the creation
of efficient motive power, could be considered as solved, since
the propulsion of the apparatus in the air would be a great deal
easier than the traction on the ground, provided that the second
part of the problem, the sustaining of the machine in the air,
would be realised.

The next day, Wednesday the 13th, no further trials were made
on account of the rain and wind.

On Thursday the 14th the Chairman requested that General
Grillon, who had just been appointed a member of the Committee,
accompany him so as to have a second witness.

The weather was fine, but a fairly strong, gusty wind was
blowing from the south. M. Ader explained to the two members
of the Committee the danger of these gusts, since at two points
of the circumference the wind would strike him sideways. The
wind was blowing in the direction A B, the apparatus starting
from C, and running in the direction shown by the arrow. The
first dangerous spot would be at B. The apparatus had been kept
in readiness in the event of the wind dying down. Toward sunset
the wind seemed to die down, as it had done on the evening of
the 12th. M. Ader hesitated, which, unfortunately, further
events only justified, but decided to make a new trial.

At the start, which took place at 5.15 p.m., the apparatus,
having the wind in the rear, seemed to run at a fairly regular
speed; it was, nevertheless, easy to note from the marks of the
wheels on the ground that the rear part of the apparatus had been
lifted and that the rear wheel, being the rudder, had not been in
constant contact with the ground. When the machine came to the
neighbourhood of B, the two members of the Committee saw the
machine swerve suddenly out of the track in a semicircle, lean
over to the right and finally stop. They immediately proceeded
to the point where the accident had taken place and endeavoured
to find an explanation for the same. The Chairman finally
decided as follows:

M. Ader was the victim of a gust of wind which he had feared as
he explained before starting out; feeling himself thrown out of
his course, he tried to use the rudder energetically, but at that
time the rear wheel was not in contact with the ground, and
therefore did not perform its function; the canvas rudder, which
had as its purpose the manoeuvring of the machine in the air, did
not have sufficient action on the ground. It would have been
possible without any doubt to react by using the propellers at
unequal speed, but M. Ader, being still inexperienced, had not
thought of this. Furthermore, he was thrown out of his course so
quickly that he decided, in order to avoid a more serious
accident, to stop both engines. This sudden stop produced the
half-circle already described and the fall of the machine on its

The damage to the machine was serious; consisting at first sight
of the rupture of both propellers, the rear left wheel and the
bending of the left wing tip. It will only be possible to
determine after the machine is taken apart whether the engine,
and more particularly the organs of transmission, have been put
out of line.

Whatever the damage may be, though comparatively easy to repair,
it will take a certain amount of time, and taking into
consideration the time of year it is evident that the tests will
have to be adjourned for the present.

As has been said in the above report, the tests, though
prematurely interrupted, have shown results of great importance,
and though the final results are hard to foresee, it would seem
advisable to continue the trials. By waiting for the return of
spring there will be plenty of time to finish the tests and it
will not be necessary to rush matters, which was a partial cause
of the accident. The Chairman of the Committee personally has
but one hope, and that is that a decision be reached accordingly.

Division General,
Chairman of the Committee,

Boulogne-sur-Seine, October 21st, 1897.

Annex to the Report of October 21st.

General Grillon, who was present at the trials of the 14th, and
who saw the report relative to what happened during that day,
made the following observations in writing, which are reproduced
herewith in quotation marks. The Chairman of the Committee does
not agree with General Grillon and he answers theseobservations
paragraph by paragraph.

1. 'If the rear wheel (there is only one of these) left but
intermittent tracks on the ground, does that prove that the
machine has a tendency to rise when running at a certain speed?'

Answer.--This does not prove anything in any way, and I was very
careful not to mention this in my report, this point being
exactly what was needed and that was not demonstrated during the
two tests made on the grounds.

'Does not this unequal pressure of the two pair of wheels on the
ground show that the centre of gravity of the apparatus is
placed too far forward and that under the impulse of the
propellers the machine has a tendency to tilt forward, due to
the resistance of the air?'

Answer.--The tendency of the apparatus to rise from the rear
when it was running with the wind seemed to be brought about by
the effects of the wind on the huge wings, having a spread of 17
metres, and I believe that when the machine would have faced the
wind the front wheels would have been lifted.

During the trials of October 12th, when a complete circuit of
the track was accomplished without incidents, as I and Lieut.
Binet witnessed, there was practically no wind. I was therefore
unable to verify whether during this circuit the two front
wheels or the rear wheel were in constant contact with the
ground, because when the trial was over it was dark (it was
5.30) and the next day it was impossible to see anything because
it had rained during the night and during Wednesday morning.
But what would prove that the rear wheel was in contact with the
ground at all times is the fact that M. Ader, though
inexperienced, did not swerve from the circular track, which
would prove that he steered pretty well with his rear
wheel--this he could not have done if he had been in the air.

In the tests of the 12th, the speed was at least as great as on
the 14th.

2. 'It would seem to me that if M. Ader thought that his rear
wheels were off the ground he should have used his canvas rudder
in order to regain his proper course; this was the best way of
causing the machine to rotate, since it would have given an
angular motion to the front axle.'

Answer.--I state in my report that the canvas rudder whose
object was the manoeuvre of the apparatus in the air could have
no effect on the apparatus on the ground, and to convince
oneself of this point it is only necessary to consider the small
surface of this canvas rudder compared with the mass to be
handled on the ground, a weight of approximately 400 kg.
According to my idea, and as I have stated in my report, M. Ader
should have steered by increasing the speed on one of his
propellers and slowing down the other. He admitted afterward
that this remark was well founded, but that he did not have time
to think of it owing to the suddenness of the accident.

3. 'When the apparatus fell on its side it was under the sole
influence of the wind, since M. Ader had stopped the machine.
Have we not a result here which will always be the same when the
machine comes to the ground, since the engines will always have
to be stopped or slowed down when coming to the ground? Here
seems to be a bad defect of the apparatus under trial.'

Answer.--I believe that the apparatus fell on its side after
coming to a stop, not on account of the wind, but because the
semicircle described was on rough ground and one of the wheels
had collapsed.
October 27th, 1897.


Specification and Claims of Wright Patent, No. 821393.
Filed March 23rd, 1903. Issued May 22nd, 1906. Expires May
22nd, 1923.

To all whom it may concern.

Be it known that we, Orville Wright and Wilbur Wright, citizens
of the United States, residing in the city of Dayton, county of
Montgomery, and State of Ohio, have invented certain new and
useful Improvements in Flying Machines, of which the following
is a specification.

Our invention relates to that class of flying-machines in which
the weight is sustained by the reactions resulting when one or
more aeroplanes are moved through the air edgewise at a small
angle of incidence, either by the application of mechanical
power or by the utilisation of the force of gravity.

The objects of our invention are to provide means for
maintaining or restoring the equilibrium or lateral balance of
the apparatus, to provide means for guiding the machine both
vertically and horizontally, and to provide a structure
combining lightness, strength, convenience of construction and
certain other advantages which will hereinafter appear.

To these ends our invention consists in certain novel features,
which we will now proceed to describe and will then particularly
point out in the claims. In the accompanying drawings, Figure I
1 is a perspective view of an apparatus embodying our invention
in one form. Fig. 2 is a plan view of the same, partly in
horizontal section and partly broken away. Fig. 3 is a side
elevation, and Figs. 4 and 5 are detail views, of one form of
flexible joint for connecting the upright standards with the

In flying machines of the character to which this invention
relates the apparatus is supported in the air by reason of the
contact between the air and the under surface of one or more
aeroplanes, the contact surface being presented at a small angle
of incidence to the air. The relative movements of the air and
aeroplane may be derived from the motion of the air in the form
of wind blowing in the direction opposite to that in which the
apparatus is travelling or by a combined downward and forward
movement of the machine, as in starting from an elevated
position or by combination of these two things, and in either
case the operation is that of a soaring-machine, while power
applied to the machine to propel it positively forward will
cause the air to support the machine in a similar manner. In
either case owing to the varying conditions to be met there are
numerous disturbing forces which tend to shift the machine from
the position which it should occupy to obtain the desired
results. It is the chief object of our invention to provide
means for remedying this difficulty, and we will now proceed to
describe the construction by means of which these results are

In the accompanying drawing we have shown an apparatus embodying
our invention in one form. In this illustrative embodiment the
machine is shown as comprising two parallel superposed
aeroplanes, 1 and 2, may be embodied in a structure having a
single aeroplane. Each aeroplane is of considerably greater width
from side to side than from front to rear. The four corners of
the upper aeroplane are indicated by the reference letters a, b,
c, and d, while the corresponding corners of the lower aeroplane
2 are indicated by the reference letters e, f, g, and h. The
marginal lines ab and ef indicate the front edges of the
aeroplanes, the lateral margins of the upper aeroplane are
indicated, respectively, by the lines ad and bc, the lateral
margins of the lower aeroplane are indicated, respectively, by
the lines eh and fg, while the rear margins of the upper and
lower aeroplanes are indicated, respectively, by the lines cd and

Before proceeding to a description of the fundamental theory of
operation of the structure we will first describe the preferred
mode of constructing the aeroplanes and those portions of the
structure which serve to connect the two aeroplanes.

Each aeroplane is formed by stretching cloth or other suitable
fabric over a frame composed of two parallel transverse spars 3,
extending from side to side of the machine, their ends being
connected by bows 4 extending from front to rear of the machine.
The front and rear spars 3 of each aeroplane are connected by a
series of parallel ribs 5, which preferably extend somewhat
beyond the rear spar, as shown. These spars, bows, and ribs are
preferably constructed of wood having the necessary strength,
combined with lightness and flexibility. Upon this framework
the cloth which forms the supporting surface of the aeroplane is
secured, the frame being enclosed in the cloth. The cloth for
each aeroplane previous to its attachment to its frame is cut on
the bias and made up into a single piece approximately the size
and shape of the aeroplane, having the threads of the fabric
arranged diagonally to the transverse spars and longitudinal
ribs, as indicated at 6 in Fig. 2. Thus the diagonal threads of
the cloth form truss systems with the spars and ribs, the threads
constituting the diagonal members. A hem is formed at the rear
edge of the cloth to receive a wire 7, which is connected to the
ends of the rear spar and supported by the rearwardly-extending
ends of the longitudinal ribs 5, thus forming a
rearwardly-extending flap or portion of the aeroplane. This
construction of the aeroplane gives a surface which has very
great strength to withstand lateral and longitudinal strains, at
the same time being capable of being bent or twisted in the
manner hereinafter described.

When two aeroplanes are employed, as in the construction
illustrated, they are connected together by upright standards 8.
These standards are substantially rigid, being preferably
constructed of wood and of equal length, equally spaced along
the front and rear edges of the aeroplane, to which they are
connected at their top and bottom ends by hinged joints or
universal joints of any suitable description. We have shown one
form of connection which may be used for this purpose in Figs. 4
and 5 of the drawings. In this construction each end of the
standard 8 has secured to it an eye 9 which engages with a hook
10, secured to a bracket plate 11, which latter plate is in
turn fastened to the spar 3. Diagonal braces or stay-wires 12
extend from each end of each standard to the opposite ends of
the adjacent standards, and as a convenient mode of attaching
these parts I have shown a hook 13 made integral with the hook
10 to receive the end of one of the stay-wires, the other
stay-wire being mounted on the hook 10. The hook 13 is shown
as bent down to retain the stay-wire in connection to it, while
the hook 10 is shown as provided with a pin 14 to hold the
staywire 12 and eye 9 in position thereon. It will be seen that
this construction forms a truss system which gives the whole
machine great transverse rigidity and strength, while at the
same time the jointed connections of the parts permit the
aeroplanes to be bent or twisted in the manner which we will now
proceed to describe.

15 indicates a rope or other flexible connection extending
lengthwise of the front of the machine above the lower
aeroplane, passing under pulleys or other suitable guides 16 at
the front corners e and f of the lower aeroplane, and extending
thence upward and rearward to the upper rear corners c and d, of
the upper aeroplane, where they are attached, as indicated at
17. To the central portion of the rope there is connected a
laterally-movable cradle 18, which forms a means for moving the
rope lengthwise in one direction or the other, the cradle being
movable toward either side of the machine. We have devised this
cradle as a convenient means for operating the rope 15, and the
machine is intended to be generally used with the operator lying
face downward on the lower aeroplane, with his head to the
front, so that the operator's body rests on the cradle, and the
cradle can be moved laterally by the movements of the operator's
body. It will be understood, however, that the rope 15 may be
manipulated in any suitable manner.

19 indicates a second rope extending transversely of the
machine along the rear edge of the body portion of the lower
aeroplane, passing under suitable pulleys or guides 20 at the
rear corners g and h of the lower aeroplane and extending thence
diagonally upward to the front corners a and b of the upper
aeroplane, where its ends are secured in any suitable manner, as
indicated at 21.

Considering the structure so far as we have now described it,
and assuming that the cradle 18 be moved to the right in Figs.
1 and 2, as indicated by the arrows applied to the cradle in
Fig. 1 and by the dotted lines in Fig. 2, it will be seen that
that portion of the rope 15 passing under the guide pulley at
the corner e and secured to the corner d will be under tension,
while slack is paid out throughout the other side or half of the
rope 15. The part of the rope 15 under tension exercises a
downward pull upon the rear upper corner d of the structure and
an upward pull upon the front lower corner e, as indicated by
the arrows. This causes the corner d to move downward and the
corner e to move upward. As the corner e moves upward it
carries the corner a upward with it, since the intermediate
standard 8 is substantially rigid and maintains an equal
distance between the corners a and e at all times. Similarly,
the standard 8, connecting the corners d and h, causes the
corner h to move downward in unison with the corner d. Since
the corner a thus moves upward and the corner h moves downward,
that portion of the rope 19 connected to the corner a will be
pulled upward through the pulley 20 at the corner h, and the
pull thus exerted on the rope 19 will pull the corner b on the
other wise of the machine downward and at the same time pull the
corner g at said other side of the machine upward. This results
in a downward movement of the corner b and an upward movement of
the corner c. Thus it results from a lateral movement of the
cradle 18 to the right in Fig. 1 that the lateral margins ad
and eh at one side of the machine are moved from their normal
positions in which they lie in the normal planes of their
respective aeroplanes, into angular relations with said normal
planes, each lateral margin on this side of the machine being
raised above said normal plane at its forward end and depressed
below said normal plane at its rear end, said lateral margins
being thus inclined upward and forward. At the same time a
reverse inclination is imparted to the lateral margins bc end fg
at the other side of the machine, their inclination being
downward and forward. These positions are indicated in dotted
lines in Fig. 1 of the drawings. A movement of the cradle 18 in
the opposite direction from its normal position will reverse the
angular inclination of the lateral margins of the aeroplanes in
an obvious manner. By reason of this construction it will be
seen that with the particular mode of construction now under
consideration it is possible to move the forward corner of the
lateral edges of the aeroplane on one side of the machine either
above or below the normal planes of the aeroplanes, a reverse
movement of the forward corners of the lateral margins on the
other side of the machine occurring simultaneously. During this
operation each aeroplane is twisted or distorted around a line
extending centrally across the same from the middle of one
lateral margin to the middle of the other lateral margin, the
twist due to the moving of the lateral margins to different
angles extending across each aeroplane from side to side, so that
each aeroplane surface is given a helicoidal warp or twist. We
prefer this construction and mode of operation for the reason
that it gives a gradually increasing angle to the body of each
aeroplane from the centre longitudinal line thereof outward to
the margin, thus giving a continuous surface on each side of the
machine, which has a gradually increasing or decreasing angle of
incidence from the centre of the machine to either side. We wish
it to be understood, however, that our invention is not limited
to this particular construction, since any construction whereby
the angular relations of the lateral margins of the aeroplanes
may be varied in opposite directions with respect to the normal
planes of said aeroplanes comes within the scope of our
invention. Furthermore, it should be understood that while the
lateral margins of the aeroplanes move to different angular
positions with respect to or above and below the normal planes of
said aeroplanes, it does not necessarily follow that these
movements bring the opposite lateral edges to different angles
respectively above and below a horizontal plane since the normal
planes of the bodies of the aeroplanes are inclined to the
horizontal when the machine is in flight, said inclination being
downward from front to rear, and while the forward corners on one
side of the machine may be depressed below the normal planes of
the bodies of the aeroplanes said depression is not necessarily
sufficient to carry them below the horizontal planes passing
through the rear corners on that side. Moreover, although we
prefer to so construct the apparatus that the movements of the
lateral margins on the opposite sides of the machine are equal in
extent and opposite m direction, yet our invention is not limited
to a construction producing this result, since it may be
desirable under certain circumstances to move the lateral margins
on one side of the machine just described without moving the
lateral margins on the other side of the machine to an equal
extent in the opposite direction. Turning now to the purpose of
this provision for moving the lateral margins of the aeroplanes
in the manner described, it should be premised that owing to
various conditions of wind pressure and other causes the body of
the machine is apt to become unbalanced laterally, one side
tending to sink and the other side tending to rise, the machine
turning around its central longitudinal axis. The provision
which we have just described enables the operator to meet this
difficulty and preserve the lateral balance of the machine.
Assuming that for some cause that side of the machine which lies
to the left of the observer in Figs. 1 and 2 has shown a
tendency to drop downward, a movement of the cradle 18 to the
right of said figures, as herein before assumed, will move the
lateral margins of the aeroplanes in the manner already
described, so that the margins ad and eh will be inclined
downward and rearward, and the lateral margins bc and fg will be
inclined upward and rearward with respect to the normal planes
of the bodies of the aeroplanes. With the parts of the machine
in this position it will be seen that the lateral margins ad
and eh present a larger angle of incidence to the resisting
air, while the lateral margins on the other side of the machine
present a smaller angle of incidence. Owing to this fact, the
side of the machine presenting the larger angle of incidence
will tend to lift or move upward, and this upward movement will
restore the lateral balance of the machine. When the other side
of the machine tends to drop, a movement of the cradle 18 in the
reverse direction will restore the machine to its normal lateral
equilibrium. Of course, the same effect will be produced in the
same way in the case of a machine employing only a single

In connection with the body of the machine as thus operated we
employ a vertical rudder or tail 22, so supported as to turn
around a vertical axis. This rudder is supported at the rear
ends on supports or arms 23, pivoted at their forward ends to
the rear margins of the upper and lower aeroplanes, respectively.
These supports are preferably V-shaped, as shown, so that their
forward ends are comparatively widely separated, their pivots
being indicated at 24. Said supports are free to swing upward at
their free rear ends, as indicated in dotted lines in Fig. 3,
their downward movement being limited in any suitable manner.
The vertical pivots of the rudder 22 are indicated at 25, and one
of these pivots has mounted thereon a sheave or pulley 26, around
which passes a tiller-rope 27, the ends of which are extended out
laterally and secured to the rope 19 on opposite sides of the
central point of said rope. By reason of this construction the
lateral shifting of the cradle 18 serves to turn the rudder to
one side or the other of the line of flight. It will be observed
in this connection that the construction is such that the rudder
will always be so turned as to present its resisting surface on
that side of the machine on which the lateral margins of the
aeroplanes present the least angle of resistance. The reason of
this construction is that when the lateral margins of the
aeroplanes are so turned in the manner hereinbefore described as
to present different angles of incidence to the atmosphere, that
side presenting the largest angle of incidence, although being
lifted or moved upward in the manner already described, at the
same time meets with an increased resistance to its forward
motion, while at the same time the other side of the machine,
presenting a smaller angle of incidence, meets with less
resistance to its forward motion and tends to move forward more
rapidly than the retarded side. This gives the machine a
tendency to turn around its vertical axis, and this tendency if
not properly met will not only change the direction of the front
of the machine, but will ultimately permit one side thereof to
drop into a position vertically below the other side with the
aero planes in vertical position, thus causing the machine to
fall. The movement of the rudder, hereinbefore described,
prevents this action, since it exerts a retarding influence on

Facebook Google Reddit Twitter Pinterest