The legend of Parson Rudall and the Botathen Ghost will be recognised by many Cornish people as a local remembrance of their boyhood. It appears from the diary of this learned master of the grammar-school--for such was his office, as ... Read more of The Botathen Ghost at Scary Stories.caInformational Site Network Informational
Privacy

THE REAL FLYING MACHINE.






From: Flying Machines Construction Operation

We will now assume that you have become proficient
enough to warrant an attempt at the construction of a
real flying machine--one that will not only remain suspended
in the air at the will of the operator, but make
respectable progress in whatever direction he may desire to go.
The glider, it must be remembered, is not
steerable, except to a limited extent, and moves only in
one direction--against the wind. Besides this its power
of flotation--suspension in the air--is circumscribed.

Larger Surface Area Required.

The real flying machine is the glider enlarged, and
equipped with motor and propeller. The first thing to
do is to decide upon the size required. While a glider
of 20 foot spread is large enough to sustain a man it
could not under any possible conditions, be made to rise
with the weight of the motor, propeller and similar
equipment added. As the load is increased so must the
surface area of the planes be increased. Just what this
increase in surface area should be is problematical as
experienced aviators disagree, but as a general proposition
it may be placed at from three to four times the area of
a 20-foot glider.[3]

[3] See chapter XXV.

Some Practical Examples.

The Wrights used a biplane 41 feet in spread, and 6 1/2
ft. deep. This, for the two planes, gives a total surface
area of 538 square feet, inclusive of auxiliary planes.
This sustains the engine equipment, operator, etc., a total
weight officially announced at 1,070 pounds. It shows
a lifting capacity of about two pounds to the square
foot of plane surface, as against a lifting capacity of
about 1/2 pound per square foot of plane surface for the
20-foot glider. This same Wright machine is also reported
to have made a successful flight, carrying a total
load of 1,100 pounds, which would be over two pounds
for each square foot of surface area, which, with auxiliary
planes, is 538 square feet.

To attain the same results in a monoplane, the single
surface would have to be 60 feet in spread and 9 feet
deep. But, while this is the mathematical rule, Bleriot
has demonstrated that it does not always hold good.
On his record-breaking trip across the English channel,
July 25th, 1909, the Frenchman was carried in a
monoplane 24 1/2 feet in spread, and with a total sustaining
surface of 150 1/2 square feet. The total weight of
the outfit, including machine, operator and fuel sufficient
for a three-hour run, was only 660 pounds. With
an engine of (nominally) 25 horsepower the distance of
21 miles was covered in 37 minutes.

Which is the Best?

Right here an established mathematical quantity is
involved. A small plane surface offers less resistance
to the air than a large one and consequently can attain
a higher rate of speed. As explained further on in this
chapter speed is an important factor in the matter of
weight-sustaining capacity. A machine that travels one-
third faster than another can get along with one-half the
surface area of the latter without affecting the load. See
the closing paragraph of this chapter on this point. In
theory the construction is also the simplest, but this is
not always found to be so in practice. The designing
and carrying into execution of plans for an extensive
area like that of a monoplane involves great skill and
cleverness in getting a framework that will be strong
enough to furnish the requisite support without an undue excess
of weight. This proposition is greatly simplified
in the biplane and, while the speed attained by the latter
may not be quite so great as that of the monoplane, it
has much larger weight-carrying capacity.

Proper Sizes For Frame.

Allowing that the biplane form is selected the construction
may be practically identical with that of the
20-foot glider described in chapter V., except as to size
and elimination of the armpieces. In size the surface
planes should be about twice as large as those of the
20-foot glider, viz: 40 feet spread instead of 20, and 6 feet
deep instead of 3. The horizontal beams, struts, stanchions,
ribs, etc., should also be increased in size proportionately.

While care in the selection of clear, straight-grained
timber is important in the glider, it is still more important
in the construction of a motor-equipped flying
machine as the strain on the various parts will be much
greater.

How to Splice Timbers.

It is practically certain that you will have to resort to
splicing the horizontal beams as it will be difficult, if not
impossible, to find 40-foot pieces of timber totally free
from knots and worm holes, and of straight grain.

If splicing is necessary select two good 20-foot pieces,
3 inches wide and 1 1/2 inches thick, and one 10-foot long,
of the same thickness and width. Plane off the bottom
sides of the 10-foot strip, beginning about two feet back
from each end, and taper them so the strip will be about
3/4 inch thick at the extreme ends. Lay the two 20-foot
beams end to end, and under the joint thus made place
the 10-foot strip, with the planed-off ends downward.
The joint of the 20-foot pieces should be directly in the
center of the 10-foot piece. Bore ten holes (with a 1/4-
inch augur) equi-distant apart through the 20-foot
strips and the 10-foot strip under them. Through these
holes run 1/4-inch stove bolts with round, beveled heads.
In placing these bolts use washers top and bottom, one
between the head and the top beam, and the other between
the bottom beam and the screw nut which holds
the bolt. Screw the nuts down hard so as to bring the
two beams tightly together, and you will have a rigid
40-foot beam.

Splicing with Metal Sleeves.

An even better way of making a splice is by tonguing
and grooving the ends of the frame pieces and enclosing
them in a metal sleeve, but it requires more mechanical
skill than the method first named. The operation of
tonguing and grooving is especially delicate and calls
for extreme nicety of touch in the handling of tools, but
if this dexterity is possessed the job will be much more
satisfactory than one done with a third timber.

As the frame pieces are generally about 1 1/2 inch in
diameter, the tongue and the groove into which the
tongue fits must be correspondingly small. Begin by
sawing into one side of one of the frame pieces about 4
inches back from the end. Make the cut about 1/2 inch
deep. Then turn the piece over and duplicate the cut.
Next saw down from the end to these cuts. When the
sawed-out parts are removed you will have a "tongue"
in the end of the frame timber 4 inches long and 1/2 inch
thick. The next move is to saw out a 5/8-inch groove in
the end of the frame piece which is to be joined. You
will have to use a small chisel to remove the 5/8-inch bit.
This will leave a groove into which the tongue will fit
easily.

Joining the Two Pieces.

Take a thin metal sleeve--this is merely a hollow tube
of aluminum or brass open at each end--8 inches long,
and slip it over either the tongued or grooved end of one
of the frame timbers. It is well to have the sleeve fit
snugly, and this may necessitate a sand-papering of the
frame pieces so the sleeve will slip on.

Push the sleeve well back out of the way. Cover the
tongue thoroughly with glue, and also put some on the
inside of the groove. Use plenty of glue. Now press
the tongue into the groove, and keep the ends firmly
together until the glue is thoroughly dried. Rub off the
joint lightly with sand-paper to remove any of the glue
which may have oozed out, and slip the sleeve into place
over the joint. Tack the sleeve in position with small
copper tacks, and you will have an ideal splice.

The same operation is to be repeated on each of the
four frame pieces. Two 20-foot pieces joined in this
way will give a substantial frame, but when suitable
timber of this kind can not be had, three pieces, each 6
feet 11 inches long, may be used. This would give 20
feet 9 inches, of which 8 inches will be taken up in the
two joints, leaving the frame 20 feet 1 inch long.

Installation of Motor.

Next comes the installation of the motor. The kinds
and efficiency of the various types are described in the
following chapter (IX). All we are interested in at
this point is the manner of installation. This varies
according to the personal ideas of the aviator. Thus one
man puts his motor in the front of his machine, another
places it in the center, and still another finds the rear of
the frame the best. All get good results, the comparative
advantages of which it is difficult to estimate. Where
one man, as already explained, flies faster than another,
the one beaten from the speed standpoint has an advantage
in the matter of carrying weight, etc.

The ideas of various well-known aviators as to the
correct placing of motors may be had from the following:

Wrights--In rear of machine and to one side.

Curtiss--Well to rear, about midway between upper
and lower planes.

Raich--In rear, above the center.

Brauner-Smith--In exact center of machine.

Van Anden--In center.

Herring-Burgess--Directly behind operator.

Voisin--In rear, and on lower plane.

Bleriot--In front.

R. E. P.--In front.

The One Chief Object.

An even distribution of the load so as to assist in
maintaining the equilibrium of the machine, should be
the one chief object in deciding upon the location of the
motor. It matters little what particular spot is selected
so long as the weight does not tend to overbalance the
machine, or to "throw it off an even keel." It is just
like loading a vessel, an operation in which the expert
seeks to so distribute the weight of the cargo as to keep
the vessel in a perfectly upright position, and prevent a
"list" or leaning to one side. The more evenly the cargo
is distributed the more perfect will be the equilibrium of
the vessel and the better it can be handled. Sometimes,
when not properly stowed, the cargo shifts, and this at
once affects the position of the craft. When a ship
"lists" to starboard or port a preponderating weight of
the cargo has shifted sideways; if bow or stern is unduly
depressed it is a sure indication that the cargo has shifted
accordingly. In either event the handling of the craft
becomes not only difficult, but extremely hazardous.
Exactly the same conditions prevail in the handling of a
flying machine.

Shape of Machine a Factor.

In placing the motor you must be governed largely by
the shape and construction of the flying machine frame.
If the bulk of the weight of the machine and auxiliaries
is toward the rear, then the natural location for the motor
will be well to the front so as to counterbalance the
excess in rear weight. In the same way if the
preponderance of the weight is forward, then the motor
should be placed back of the center.

As the propeller blade is really an integral part of the
motor, the latter being useless without it, its placing
naturally depends upon the location selected for the
motor.

Rudders and Auxiliary Planes.

Here again there is great diversity of opinion among
aviators as to size, location and form. The striking
difference of ideas in this respect is well illustrated in
the choice made by prominent makers as follows:

Voisin--horizontal rudder, with two wing-like planes,
in front; box-like longitudinal stability plane in rear,
inside of which is a vertical rudder.

Wright--large biplane horizontal rudder in front at
considerable distance--about 10 feet--from the main
planes; vertical biplane rudder in rear; ends of upper
and lower main planes made flexible so they may be
moved.

Curtiss--horizontal biplane rudder, with vertical damping
plane between the rudder planes about 10 feet in
front of main planes; vertical rudder in rear; stabilizing
planes at each end of upper main plane.

Bleriot--V-shaped stabilizing fin, projecting from rear
of plane, with broad end outward; to the broad end of
this fin is hinged a vertical rudder; horizontal biplane
rudder, also in rear, under the fin.

These instances show forcefully the wide diversity of
opinion existing among experienced aviators as to the
best manner of placing the rudders and stabilizing, or
auxiliary planes, and make manifest how hopeless would
be the task of attempting to select any one form and
advise its exclusive use.

Rudder and Auxiliary Construction.

The material used in the construction of the rudders
and auxiliary planes is the same as that used in the main
planes--spruce for the framework and some kind of
rubberized or varnished cloth for the covering. The
frames are joined and wired in exactly the same manner
as the frames of the main planes, the purpose being to
secure the same strength and rigidity. Dimensions of
the various parts depend upon the plan adopted and the
size of the main plane.

No details as to exact dimensions of these rudders and
auxiliary planes are obtainable. The various builders,
while willing enough to supply data as to the general
measurements, weight, power, etc., of their machines,
appear to have overlooked the details of the auxiliary
parts, thinking, perhaps, that these were of no particular
import to the general public. In the Wright machine, the
rear horizontal and front vertical rudders may be set
down as being about one-quarter (probably a little less)
the size of the main supporting planes.

Arrangement of Alighting Gear.

Most modern machines are equipped with an alighting
gear, which not only serves to protect the machine and
aviator from shock or injury in touching the ground, but
also aids in getting under headway. All the leading
makes, with the exception of the Wright, are furnished
with a frame carrying from two to five pneumatic rubber-
tired bicycle wheels. In the Curtiss and Voisin
machines one wheel is placed in front and two in the
rear. In the Bleriot and other prominent machines the
reverse is the rule--two wheels in front and one in the
rear. Farman makes use of five wheels, one in the,
extreme rear, and four, arranged in pairs, a little to the
front of the center of the main lower plane.

In place of wheels the Wright machine is equipped
with a skid-like device consisting of two long beams
attached to the lower plane by stanchions and curving
up far in front, so as to act as supports to the horizontal
rudder.

Why Wood Is Favored.

A frequently asked question is: "Why is not aluminum,
or some similar metal, substituted for wood."
Wood, particularly spruce, is preferred because, weight
considered, it is much stronger than aluminum, and this
is the lightest of all metals. In this connection the following
table will be of interest:

Compressive
Weight Tensile Strength Strength
per cubic foot per sq. inch per sq. inch
Material in lbs. in lbs. in lbs.
Spruce . . . . 25 8,000 5,000
Aluminum 162 16,000 ......
Brass (sheet) 510 23,000 12,000
Steel (tool) 490 100,000 40,000
Copper (sheet) 548 30,000 40,000

As extreme lightness, combined with strength,
especially tensile strength, is the great essential in flying-
machine construction, it can be readily seen that the
use of metal, even aluminum, for the framework, is
prohibited by its weight. While aluminum has double the
strength of spruce wood it is vastly heavier, and thus
the advantage it has in strength is overbalanced many
times by its weight. The specific gravity of aluminum
is 2.50; that of spruce is only 0.403.

Things to Be Considered.

In laying out plans for a flying machine there are five
important points which should be settled upon before
the actual work of construction is started. These are:

First--Approximate weight of the machine when finished
and equipped.

Second--Area of the supporting surface required.

Third--Amount of power that will be necessary to
secure the desired speed and lifting capacity.

Fourth--Exact dimensions of the main framework
and of the auxiliary parts.

Fifth--Size, speed and character of the propeller.

In deciding upon these it will be well to take into
consideration the experience of expert aviators regarding
these features as given elsewhere. (See chapter X.)

Estimating the Weights Involved.

In fixing upon the probable approximate weight in
advance of construction much, of course, must be assumed.
This means that it will be a matter of advance
estimating. If a two-passenger machine is to be built
we will start by assuming the maximum combined
weight of the two people to be 350 pounds. Most of
the professional aviators are lighter than this. Taking
the medium between the weights of the Curtiss and
Wright machines we have a net average of 850 pounds
for the framework, motor, propeller, etc. This, with
the two passengers, amounts to 1,190 pounds. As the
machines quoted are in successful operation it will be
reasonable to assume that this will be a safe basis to
operate on.

What the Novice Must Avoid.

This does not mean, however, that it will be safe to
follow these weights exactly in construction, but that
they will serve merely as a basis to start from. Because
an expert can turn out a machine, thoroughly equipped,
of 850 pounds weight, it does not follow that a novice
can do the same thing. The expert's work is the result
of years of experience, and he has learned how to construct
frames and motor plants of the utmost lightness
and strength.

It will be safer for the novice to assume that he can
not duplicate the work of such men as Wright and Curtiss
without adding materially to the gross weight of
the framework and equipment minus passengers.

How to Distribute the Weight.

Let us take 1,030 pounds as the net weight of the machine
as against the same average in the Wright and
Curtiss machines. Now comes the question of distributing
this weight between the framework, motor, and
other equipment. As a general proposition the framework
should weigh about twice as much as the complete
power plant (this is for amateur work).

The word "framework" indicates not only the wooden
frames of the main planes, auxiliary planes, rudders,
etc., but the cloth coverings as well--everything in fact
except the engine and propeller.

On the basis named the framework would weigh 686
pounds, and the power plant 344. These figures are
liberal, and the results desired may be obtained well
within them as the novice will learn as he makes progress
in the work.

Figuring on Surface Area.

It was Prof. Langley who first brought into prominence
in connection with flying machine construction the
mathematical principle that the larger the object the
smaller may be the relative area of support. As explained
in chapter XIII, there are mechanical limits as
to size which it is not practical to exceed, but the main
principle remains in effect.

Take two aeroplanes of marked difference in area of
surface. The larger will, as a rule, sustain a greater
weight in relative proportion to its area than the smaller
one, and do the work with less relative horsepower. As
a general thing well-constructed machines will average
a supporting capacity of one pound for every one-half
square foot of surface area. Accepting this as a working
rule we find that to sustain a weight of 1,200 pounds
--machine and two passengers--we should have 600
square feet of surface.

Distributing the Surface Area.

The largest surfaces now in use are those of the
Wright, Voisin and Antoinette machines--538 square
feet in each. The actual sustaining power of these machines,
so far as known, has never been tested to the
limit; it is probable that the maximum is considerably
in excess of what they have been called upon to show.
In actual practice the average is a little over one pound
for each one-half square foot of surface area.

Allowing that 600 square feet of surface will be used,
the next question is how to distribute it to the best
advantage. This is another important matter in which
individual preference must rule. We have seen how
the professionals disagree on this point, some using
auxiliary planes of large size, and others depending upon
smaller auxiliaries with an increase in number so as to
secure on a different plan virtually the same amount of
surface.

In deciding upon this feature the best thing to do is
to follow the plans of some successful aviator, increasing
the area of the auxiliaries in proportion to the increase
in the area of the main planes. Thus, if you use 600
square feet of surface where the man whose plans you
are following uses 500, it is simply a matter of making
your planes one-fifth larger all around.

The Cost of Production.

Cost of production will be of interest to the amateur
who essays to construct a flying machine. Assuming
that the size decided upon is double that of the glider
the material for the framework, timber, cloth, wire, etc.,
will cost a little more than double. This is because it
must be heavier in proportion to the increased size of
the framework, and heavy material brings a larger price
than the lighter goods. If we allow $20 as the cost of
the glider material it will be safe to put down the cost
of that required for a real flying machine framework
at $60, provided the owner builds it himself.

As regards the cost of motor and similar equipment
it can only be said that this depends upon the selection
made. There are some reliable aviation motors which
may be had as low as $500, and there are others which
cost as much as $2,000.

Services of Expert Necessary.

No matter what kind of a motor may be selected the
services of an expert will be necessary in its proper
installation unless the amateur has considerable genius
in this line himself. As a general thing $25 should be
a liberal allowance for this work. No matter how carefully
the engine may be placed and connected it will be
largely a matter of luck if it is installed in exactly the
proper manner at the first attempt. The chances are
that several alterations, prompted by the results of trials,
will have to be made. If this is the case the expert's bill may
readily run up to $50. If the amateur is competent to do this
part of the work the entire item of $50 may, of course, be cut
out.

As a general proposition a fairly satisfactory flying machine,
one that will actually fly and carry the operator with it, may be
constructed for $750, but it will lack the better qualities which
mark the higher priced machines. This computation is made on
the basis of $60 for material, $50 for services of expert, $600
for motor, etc., and an allowance of $40 for extras.

No man who has the flying machine germ in his system will be long
satisfied with his first moderate price machine, no matter how
well it may work. It's the old story of the automobile "bug"
over again. The man who starts in with a modest $1,000 automobile
invariably progresses by easy stages to the $4,000 or $5,000
class. The natural tendency is to want the biggest and best
attainable within the financial reach of the owner.

It's exactly the same way with the flying machine
convert. The more proficient he becomes in the manipulation
of his car, the stronger becomes the desire to fly
further and stay in the air longer than the rest of his
brethren. This necessitates larger, more powerful, and
more expensive machines as the work of the germ progresses.

Speed Affects Weight Capacity.

Don't overlook the fact that the greater speed you
can attain the smaller will be the surface area you can
get along with. If a machine with 500 square feet of
sustaining surface, traveling at a speed of 40 miles an
hour, will carry a weight of 1,200 pounds, we can cut
the sustaining surface in half and get along with 250
square feet, provided a speed of 60 miles an hour can
be obtained. At 100 miles an hour only 80 square feet
of surface area would be required. In both instances the
weight sustaining capacity will remain the same as with
the 500 square feet of surface area--1,200 pounds.

One of these days some mathematical genius will
figure out this problem with exactitude and we will have
a dependable table giving the maximum carrying capacity
of various surface areas at various stated speeds,
based on the dimensions of the advancing edges. At
present it is largely a matter of guesswork so far as
making accurate computation goes. Much depends upon
the shape of the machine, and the amount of surface
offering resistance to the wind, etc.





Next: SELECTION OF THE MOTOR.

Previous: PUTTING ON THE RUDDER.



Add to del.icio.us Add to Reddit Add to Digg Add to Del.icio.us Add to Google Add to Twitter Add to Stumble Upon
Add to Informational Site Network
Report
Privacy
SHAREADD TO EBOOK


Viewed: 1790