es09Scr00Xx

WM

|
AW
AT
LYVYGIT VINIDYIA JO ALISMAAINObs, 6/2: D46

  
  
 
  
  
   
 
 
 
 
 
 
 
 
 
 
  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
    
 
 
 
     
 
    
     

PROPOSED HANDBOOK FOR AIRPLANE DESIGNERS

CONTENTS
Page
Me FCN Ol Bbc. ec pew oo g vk ne ee Ge BG ek Jia ae Sas ee eee ee 1
Se 41 —-tructiral desmn requiremelite....- 2. acs tt Lie once 4
meee. 1. oauired loads for airplanes (i202 eu Les A ee eee 4
. watt o... Mideellancous feduirémoents... . 24cbes ec- Cie ib oiciie 2k. eo a OR ae 21
Memes 011 Matorigle i. ta oh ch mice wre a bmeagrh ouean ks dam yeaa aes Pee eee 28

SECTION I.—_GENERAL

1. The following pages are issued as a preliminary draft of the Department of Com-
merce requirements as to methods of distributing the loads and analyzing the structures
of airplanes submitted to the department for approval. They are intended to supplement
the Air Commerce Regulations, effective December 31, 1926.

2. It is appreciated by the department that with a specification covering so broad a
field and with the developments constantly being made in the science of aviation, situations
‘will arise where deviations from the methods described here will be necessary. In such

eases the department will consider granting the manufacturer or designer a deviation from
one or several of the rules if sufficient evidence is submitted to enable the department’s
Fengineers to determine whether or not the proposed deviation will adversely affect the J
airworthiness of the design.
These rules are based on some 10 years’ development in the science of airplane struc-
F tural design, and experience indicates that they will give a safe and well-proportioned
structure if carefully followed.
3. In determining the airworthiness of a given airplane the following factors are taken
= into consideration:
E (a) The structural strength of wings, ailerons, tail surfaces, fuselage, including engine
fmount, fittings, control system, and landing gear.
. (b) Cockpit, cabin, and control arrangements.
(c) Power plant and power-plant installation.
(d) Equipment and instruments.
(e) Propellers.
(f) Design of fittings.
(g) Materials and workmanship.
(h) Flying characteristics and qualities.
= Certain of these items may be adequately demonstrated by theoretical analyses and
= drawings, others by visual inspection, and still others by tests. Of the items demonstrable
| by analyses and drawings those listed in subparagraph (a) are of the utmost importance
as affecting the structural safety of the design. The rules given in the following pages
| describe methods which are approved by the department for use in determining the
estrength of the main structural members of an airplane.
= 4. The procedure followed by the department when approving designs for a manu-
= facturer’s approved type certificate is as follows:
The drawings, which the manufacturer is required to furnish in duplicate, are examined
efor deviations from what is considered good practice and they are then checked against
= the stress analyses required by part 4, subparagraph (0) of section 21 of the Regulations
to ascertain whether or not the-structure investigated conforms with that to be built.

(1)

 

   

Oct 17492

The stress analysis is checked for errors in assumptions, for deviations from the
approved methods, and for errors in arithmetic. If the strength of the structure is shown
by the analyses to meet the requirements listed in part 1 of Section II of this handbook and
if no deviations from what is commonly accepted as good practice are noted, the structurg
is approved.

One set of the drawings and analyses is then impressed with the seal of the Depart-
ment of Commerce and returned to the manufacturer to be used in the construction of his
airplanes. The duplicate copy is placed in the department’s files. Airplanes built to
conform exactly to these approved drawings may, upon the manufacturer making affidavit
to that fact, be accepted as structurally airworthy. Any deviations must, however, be
approved by the department before being incorporated in the construction of any airplane.

The procedure followed in the case of licensing single airplanes of any type will be
similar.

5. It is obvious that the foregoing procedure places upon the manufacturer the burden
of proving that his design is structurally safe, and he should keep this in mind when sub-
mitting data to the department. In the majority of cases the department’s engineers will
not be able to see an airplane of the type they are examining, at the time their check is
made, and the manufacturer must therefore make his drawings and analyses as clear as
possible if he wishes to avoid delay and controversy in obtaining an approval. Mere
statements of opinion should never be included in an analysis without presenting the facts
upon which the opinion is based. Deviations made from the specified methods of analysis
because the structure being investigated does not conform with the conventional, or because
the manufacturer has data indicating that his method is safe, must be accompanied by a
clear explanation of the reasons for the deviation and supported by any data which may
be available.

6. In order that the structure may be shown in sufficient detail to be used for the
construction of the airplanes to be built under an approved type certificate, the drawings
must be quite complete and well dimensioned. Assembly drawings of the major struc-
tural units, such as wings, stabilizer, elevator, etc., will suffice if they are completely
dimensioned and if they show the cross sections of all wooden members, or metal members
of special design, and the sizes of standard wires, tubes, etc., used in the assembly. The
location of hinges, control masts, and the points of attachment of all brace struts or wires
should be clearly shown. Drawings should be made to scale and important dimensions
given. Dimensions referring to lengths of members, distances between supports, etc., are
more satisfactory if shown in inch units rather than feet and inches.

In order to represent the structure satisfactorily, the drawings submitted should
include:

(1) A general assembly of each wing with the sizes and locations of spars, drag struts,
drag wires, leading edge, trailing edge, and members supporting the ailerons and controls
clearly. indicated.

f (2) Drawings of a typical rib showing method of attachment to the spars and leading
edge.

(3) Line diagrams of the external wing bracing showing truss dimensions, sizes of
struts, and wires.

(4) A layout of the fuselage structure showing sizes of all members of the primary
structure.

(5) Drawings of the tail surfaces showing sizes of spars, torque tubes, internal and
external braces, location of hinges, and masts.

(6) Chassis drawings giving a layout of the landing gear that shows sizes of struts,
axle, bracing wires, shock absorber spindles, spreader tubes and other important members,
and detail drawings or sketches showing the arrangement of members at their lower ends
and the method of connecting them. Drawings of the tail skid showing the skid, its
connection to the main fuselage structure, and the method of steering it, if any.3

(7) For seaplanes, drawings of the floats. showing their lines, general dimensions, and
a layout showing sizes of struts and wires and method of attachment to the fuselage.

(8) Detail drawings of the mechanisms used to adjust the stabilizer, fin, wing flaps,
or similar surfaces while in flight are also desired.

(9) Drawings showing the main wing fittings and the control system are required.
| The analyses must cover the investigation of the strength of the main members of the
wings, tail surfaces, landing gear, including tail skids, fuselage, fittings, and control systems
for the load factors required in paragraph 2, part 1, Section II. The analysis of secondary
members carrying heavy loads and the investigation of main members subjected to eccen-
tric loads are required.
| Stress analyses of the control surfaces and control systems may be supplanted by
tests made on these units, assembled as in flight, if the tests show that the unit will carry
125 per cent of the design load and if the stick, rudder bar, etc., withstand the loads speci-
fied in paragraph 5d, part 2, Section II. It is recommended that both analyses and tests
be made on these units and on the main wing fittings, though both are not required.

It is particularly important that the stress analysis state the material used for various
members, whether or not it is heat treated, and what physical properties are guaranteed,
not assumed, by the manufacturer. If metal members of special design are used, test data
showing their strength properties under loads similar to those to which they will be
subjected in the structure must be submitted.

An analysis should be carefully checked as to arithmetic before being subniitted
for approval, as the department will return analyses for correction .when arithmetical
errors are found, except when their effect is obviously negligible.SECTION II._STRUCTURAL DESIGN REQUIREMENTS

 
  
   
  
   
  

PART 1. REQUIRED LOADS FOR AIRPLANES

1. Definitions.—The National Advisory Committee Report on Nomenclature,
No. 240, gives definitions of the words and phrases used in aeronautics but some of those
having special applications in the field of structural design are not included. That the
subsequent paragraphs of this section may be clear, the meaning of such words and
phrases, as they are commonly employed in the design of airplane structures, is given here.

Aspect ratio—The aspect ratio of a wing or control surface is the ratio of the square
of the span to the wing area. It is defined in this manner in order to make the definition
easily applicable to wings that are not rectangular in shape. In computing the aspect
ratio of a wing, part of which is covered by nacelles or a fuselage, the portion of the wing
so covered shall be included in the wing area. Such covered portions of the wings shall
not be included in the wing area, however, in any other computations.

Basic load.—The basic loads on (or stresses in) an airplane or part are the loads on
(or stresses in) the airplane or part when the airplane is at rest or in a condition of unac-
celerated flight. (Compare design load, ultimate load, load factor, factor of safety,
margin of safety, normal load.)

Beam direction..—The direction parallel to the intersection of the plane of the spar
web and the plane of symmetry. (Compare chord, drag, lift, side, and vertical directions.)

Beam force (or component).'—A force (or component) in the beam direction—that is,
parallel to the intersection of the plane of the spar web and the plane of symmetry. (Com-
pare chord, drag, lift, side, and vertical forces.)

Center of gravity —Usually refers to the center of gravity of the airplane fully loaded.
Sometimes used for the center of gravity of a single load or a special set of loads, in which
case the load or loads concerned should be stated or clearly implied by the context.

Chord direction.\—The direction of the intersection of the plane of the internal wing
truss with the plane of symmetry. When a wing has two internal trusses in nonparallel
planes, the plane bisecting the dihedral angle between those two planes should be used,
(Compare beam, drag, lift, side, and vertical directions.)

Chord force (or component).'—A force (or component) in the chord direction; that is,
parallel to the plane of the internal wing truss and to the plane of symmetry. (Compare
beam, drag, lift, side, and vertical forces.)

Design load.—The load for which a member is designed. It is usually obtained by
multiplying a basic load by a specified load factor. (Compare basic load, factor of safety,
margin of safety, normal load, ultimate load.)

Double taper (spars).—A spar is double tapered when both width and depth vary along
the span.

Double taper (wings).—A wing is double tapered when both the chord and thickness
ratio vary along the span. (Compare equivalent wing, taper in plan only, taper in thick
ness ratio only, thickness ratio.) :

Drag direction.—The direction of the relative wind. The angle between the chord and
drag directions is usually small and often zero. (Compare beam, chord, lift, side, and
vertical directions. )

 

 

 

 

 

 

 

 

 

  

 

 

 

  
  
 
 
  
 

 

 

 

 

 

 

 

1 Beam and chord directions and forces are to be distinguished from lift and drag directions and forces. The latter depend on thé
direction of the relative wind while the former depend upon the dimensions of the airplane. The beam direction and forces are so named
because forces in that direction are considered as loads to be supported by the wing spars acting as beams. The chord direction and forces}
are so named because they are approximately, at least, in the plane of the wing chord,

(4)5

4

Drag force (or component).—A force (or component) in the drag direction—that is,
parallel to the relative wind. (Compare beam, chord, lift, side, and vertical forces.)

Effective span.—The true span of a wing less corrections for tip loss, area of cut-outs,
width of nacelles, etc. For the method of computing the effective span in a given case,
see Figures 3, 4, and 5.

Efficiency of wings—The ratio of the normal gross beam load per square foot on the
upper wing to that on the lower. For the method of computation, see Figure 2.

Equivalent wing.—A wing of the same span as the actual wing but with the chord at
each section reduced in proportion to the ratio of the average beam load at that section to
the average beam load at the section taken as standard.

Factor of safety —The ratio of the design load or the ultimate load to the maximum
probableload. In airplane design the least factor of safety is usually about 2. The-term is
often used incorrectly in place of load factor. (Compare load factor and margin of safety.)

Flying conditions —The loading conditions of high incidence, low incidence, inverted
flight, nose dive, and maximum tail load.

Gap.—The distance between the planes of the chords of two adjacent wings of an
airplane, measured along a line perpendicular to the chord of the upper wing at any desig-
nated point of its leading edge.

Gross weight.—The total weight of the airplane including useful load. (Compare net
weight.)

High incidence.—A loading condition for the wings and fuselage representing the
condition of flying at a high angle of attack while pulling out of a dive. (See low incidence,
inverted flight, and nose dive.)

Inverted flight.—A loading condition for the wings representing the conditions of flying
upside down and of commencing a dive. (Compare high incidence, low incidence, and
nose dive.)

Investigation of stresses —When an investigation of the stresses in a unit in a given
condition of loading is called for, only those computations are required that are needed to
show which members, if any, should be designed for that condition and the stresses in those
members. Often an investigation will show that no members need be designed for the
loading condition considered.

Level landing.—A loading condition for the fuselage and landing gear representing
conditions in a tail-high landing.

‘ Lift direction —The direction in the plane of symmetry perpendicular to the relative
wind. (Compare beam, chord, drag, side, and vertical directions.)

Lift force (or component).—A force (or component) in the lift direction. (Compare
beam, chord, drag, side, and vertical forces.)

Load factor.—The ratio of some other load to the basic load in which the relative
distribution of the forces is the same. This other load may be any of the following: The
load applied during some special maneuver, the maximum probable load on the airplane
or part, the design load, or the ultimate load. Whenever a load factor is mentioned, the
context should show clearly what load is being compared to the basic load. (Compare
basic load, design load, factor of safety, margin of safety, and ultimate load.)

Low incidence.—A loading condition for the wings and fuselage, corresponding to a
condition of flying with a small angle of attack while pulling out of a dive, and designed to
insure adequate strength in some members that are lightly loaded in the other loading
conditions. (Compare high incidence, inverted flight, nose dive, and maximum tail load
condition. )

Margin of safety—The ratio of the ultimate strength of a member minus the design
load to the design load. Usually a member with a positive margin of safety is satisfactory
and one with a negative margin of safety is unsatisfactory. (Compare design load, factor
of safety, load factor, and ultimate load.)6

Maximum tail load conditions—Loading conditions for the tail surfaces and fuselage
which represent conditions when the loads on the tail surfaces are at a maximum.

Minimum size.—The smallest sized member that should be used in a given location
as determined by considerations of maintenance, loads in handling, and similar practical

 

considerations. ; :
Net weight—The gross weight, less some specified partial weight. Very often the

partial weight is the dead weight of the wings, or it may be the useful load. The partial
weight in question should always be clearly indicated in the context. ;

Normal load.—The load on that part of a wing assumed to be unaffected by tip losses
or similar corrections. In any given case it may be a basic, design, gross, net, or ultimate
load, depending on the context. 5

Nose dive.—A loading condition on the wings representing the conditions in a dive at
the angle of attack of zero lift. (Compare high incidence, low incidence, and inverted
flight. ) :

Primary structure—Each structural member, the failure of which would seriously
impair the safety of the airplane, is considered a part of the primary structure.

Restraint coefficient —A coefficient used in column formulas to represent the degree of
restraint of the ends of the column. For pin ended columns c=1, and for rigidly fixed
ended columns c=4. In airplane work it is very seldom advisable to use a value of ¢
greater than 2.

Side direction.—The direction perpendicular to the plane of symmetry. (Compare
beam, chord, drag, lift, and vertical directions.)

Side force (or component).—A force (or component) perpendicular to the plane of sym-
metry. (Compare beam, chord, drag, lift, and vertical forces. )

Stagger.—The angle between a line joining the forward third points of the wings and a
line perpendicular to the chord of the upper wing, both lines being drawn in a plane parallel
to the plane of symmetry. Stagger is considered positive if the forward third point of the
upper wing is in advance of that of the lower. Sometimes stagger is expressed in per cent of
gap, in which case the per cent of stagger is equal to the tangent of the angle of stagger.
It is indicated by @ in Figure 2.

Taper in plan only.—A wing is ‘‘tapered in plan only” when the wing chord varies
along the span but all cross sections of the wing are geometrically similar. (Compare
double taper, taper in thickness ratio only, and equivalent wing.)

Taper in thickness ratio only.—A wing is ‘‘tapered in thickness ratio only’’ when the
wing chord is constant but the cross sections are not geometrically similar. (Compare
double taper, taper in plan only, equivalent wing, and thickness ratio.)

Thickness ratio.—The ratio of the maximum thickness of an airfoil section to its
chord. (Compare taper in thickness ratio only and equivalent wing.)

- Three-point landing.—A loading condition for the fuselage and landing gear represent
ing the maneuver of landing with wheels and tail skid touching the ground simultaneously.
(Compare level landing.)

Ultimate load.—The load that caused failure in a test or the load that, according to
computations, should cause failure in a member. (Compare basic load, design load,
normal load, factor of safety, margin of safety, and load factor.)

Ultimate load factor. —The ratio of the ultimate load to the basic load in the same load-
ing condition. (Compare basic load and ultimate load.)

Useful load.—The crew and passengers, oil, and fuel, ballast other than emergency,
and portable equipment.

Vertical direction.—In airplane structural work this is usually assumed to be in the
plane of symmetry and normal to the propeller axis. Sometimes, as in the three-point
landing condition, it may be assumed normal to some other line, such as the ground line,
In any case the context should always show what direction is assumed to be vertical.
(Compare beam, chord, drag, lift, and side directions.)

 

 

)7

Vertical force (or component).—A force (or component) in the vertical direction. (Com-
pare beam, chord, drag, lift, and side forces.)

Wing tip.—A line parallel to the plane of symmetry and tangent to the end of the
wing. (See fig. 3.)

2. Load factors.—To define the structural strength requirements for commercial
airplanes they are divided into two types. The requirements for racers or airplanes that
are to be used for acrobatics may be obtained from the department to cover any specific
design.

The primary members of the airplane structure shall be, as when new, of sufficient
strength to withstand the stresses in them for the conditions listed below.

In high incidence the load factor shall be obtained from F igure 1 for the appropriate
weight and horsepower loading. In using the figure the upper part will apply to open
cockpit airplanes having a gross weight of 3,000 pounds or less. For closed cabin and
heavy transport airplanes the load factor will be found by interpolation between the
curves for 2,500 and 12,500 pounds, the value used being dependent on the gross weight
of the airplane. For instance, a 3,600-pound airplane having two 150-horsepower engines
will have a high incidence factor of soma aoG 7-9 6.0) + 6.0=6.44, say 6.5.

In low incidence the load factor will be 65 per cent of the high incidence factor but in
no case less than 3. In inverted flight and nose dive the factor will be 40 per cent of the
high incidence factor but in no case shall they be less than 2. The nose dive condition
will be found described more fully in paragraph 14 of this section.

Table 1 gives the load factors for the landing conditions, the height of free drop required
for landing gear design, and the required design load for the tail surfaces.

TaBLE 1.—Load factors

 

: Loads per
Height of square foot,

 

 

 

|
3 | Landing | fee drop | "a ierons :
Class | eeniditions = ee RAdch oes Vertical tail surfaces
| SOLE zontal tail
| (inches) | “surfaces
Open ‘cockpit, 3,000 pounds-or less... -.... 23 | - 625 24 30
Closed cabin:
2,600; pOUNndS-OF TOSS. k Biss oo ee a ie or ed 6.0 22 25 |\75 per cent of load required on hori-
DOUG DOUNGSE eA ka Se as Se Pa ere te 5.5 | 20 20 zontal surfaces.
LD OOUSDOUNGS ee cee ea ee eR a 5.0 18 15
Over £2500 pounds. See eS es ea ee 4.5 18 16

 

 

 

For seaplanes the load factor in landing shall be 8.

The chassis shock absorber shall be -designed to absorb the energy corresponding to
the free drop listed above without subjecting the chassis to forces greater than those corre-
sponding to the load factors in landing. The tail-skid shock absorber shall be designed to
resist a free drop in the three-point landing attitude equal to that listed above without
imposing loads on the skid or fuselage greater than those corresponding to the load factors
for landing. If a shock absorber designed to dissipate the energy of the free drop by the
flow of a liquid through an orifice—the oleo or oleo-pneumatic type of absorber, is used—the
load factors required in the landing conditions may be reduced not to exceed 25 per cent.
When this type of shock absorber is used suitable provision must be made to carry the
shocks due to taxiing after the shock absorber has been forced to the full extent of its
travel. The method to be used must be clearly shown or described on the chassis drawings.
No reduction may be made on the required height of free drop for the shock absorber,

3. Loads on wings.—The beam and chord components of the loads on the wings
for the high and low incidence and inverted flight conditions shall be computed according
to the rules given in this paragraph. The principal variations for use with structures
varying from the conventional biplane are given in the succeeding paragraphs.

61109—27——-28

(a) The computations should be made in the following order:

I. Compute or estimate gross weight of airplane.
II. Compute effective span of each wing.
III. Compute or estimate dead weight of wings.
IV. Compute relative efficiency of the wings.
V. Compute normal gross beam load on each wing.
VI. Compute normal net beam load on each wing.
VII. Compute normal chord load on each wing.
VIII. Compute all other values required for plotting ct
wing.
(b) The computations under (a) for the conventional ty
follows:

irves of beam and chord loads on each

pe of design shall be made as

I. The gross weight to be used in the calculations is the weight of the airplane
with full pay load. The air load on the tail surfaces should be neglected.

Il. The effective span of a wing of constant chord and section is the distance
from wing tip to wing tip, less that portion of the length, if any, that is covered by the
fuselage and nacelles, and an allowance for tip loss. The ‘‘effective semispan’’ S,, in
Figures 4 and 5, is one half of the effective span, L. The method of determining the
decrease in effective span to allow for tip loss is shown in Figure 4 for externally
braced and in Figure 5 for internally braced wings.

III. The total dead weight Wa of the wings shall be estimated or found by
weighing a cellule after construction. The dead weight per inch run, Wa, shall be
determined by dividing Wa by the sum of the effective spans of the wings.

IV. The relative efficiency of the wings shall be determined from Figure 2,
This figure gives the ratio of the normal unit load on the upper wing to that on the
lower wing as a function of the gap-chord ratio and the stagger in degrees. In com-
puting the efficiency with this chart, the gap-chord ratio and the stagger shall be
measured on a vertical section midway between the side of the fuselage and the
effective tip of the shorter wing. The same relative efficiency shall be used for all
flying conditions.

V. The normal gross loads for each wing are to be obtained from the formulas;

Wee
1 eLytL
Where W,=gross weight of airplane in pounds.
e=efficiency of upper wing from Figure 2.
L,=effective span of upper wing in inches.
L,=effective span of lower wing in inches.
W,g=normal gross beam load on lower wing in pounds per inch run.
Wye =normal gross beam load on upper wing in pounds per inch run.
W, may be taken as half the gross weight and L, and L, as the effective semispang
Sey and Sq, if desired.

VI. The normal net beam loads, w, and w, are obtained by subtracting the
dead weight of the wings, wa, from Wug, and Wig, respectively. At this point it B
advisable to check the results by computing the total net weight from the computed
net loads per inch run.

VII. The normal chord load per inch run on each wing shall be taken as the
proportion of the normal net beam load given in Table 2 in the high incidence, as li
per cent of the normal. net beam load in the low incidence condition, and as zero il
inverted flight. In the high incidence condition this load shall be assumed to ac

° toward the leading edge. In low incidence it shall be assumed to act toward thé
trailing edge.
; VI. (a2). The chord load along the wing from the plane of symmetry to the
wing tip is assumed to be constant and equal to the normal chord load—that is, itis
assumed that there is no tip loss affecting the chord load.

and Wug = CWig9

(6) The distribution of the beam load along the span is shown by curves A, B,
C, in Figures 4 and 5. Figure 4 applies to conventional, externally braced biplanes
where the outer strut point is approximately one chord length from the wing tip and
shall be used in obtaining the stresses in all members inboard of the outer strut point
for such airplanes. In computing the design loads for those sections of the spars
between the outer point of inflection of the outer bay and the wing tip, the bending
moments obtained from the loading represented by curves A, B, C, of Figure 4 shall
be increased 30 per cent. The design loads in the outer struts and outer bay lift
wires and that part of the axial loads in the spars between the outer point of inflection
and the outer strut point that is due to the loads in the lift truss shall be increased by
15 per cent. The axial loads in other parts of the lift truss need not be increased.
For externally braced monoplanes or biplanes, where the outer strut point is at a
distance materially greater than one chord length from the tip, the following method
shall be used. For the investigation of stresses inboard of the outer strut point the
load curve represented by A, B, OC, of Figure 4 shall be used. The design loads in the
outer struts or brace wires, the bending moments and shears in the spars, and that
part of the axial loads in the spars that is due to loads in the lift truss shall, for the sec-
tion between the outer point of inflection and the wing tip, be computed for the tip
loss shown by part A, B of the loading curve of Figure 5, point B, being at a point one
chord length from the tip. For internally braced wings the loading represented by the
curve A, B, C of Figure 5 shall be used, point B, for tapered wings or wings of constant
chord, being at the section where the chord equals the distance to the wing tip.

The purpose of the rules in the above subparagraph is to obtain a structure having
sufficient strength to withstand the variations in tip loss which pressure distribution tests
indicate to be liable to occur. These rules apply only to wings having square, circular,
or elliptical tips and must not be used for raked tips where the end of the wing makes an
angle of more than 15° with the plane of symmetry of the airplane. For such special con-
ditions the designer should apply to the department for a special ruling before proceeding
with the analysis of the wings.

4. Effect of cut-outs.—In computing the loads on wings with trailing edge cut-outs
the following modifications shall be made in the method outlined in paragraph 3. The
effective span of the wing with the cut-out shall be computed as before, except that an
amount equal to the area of the cut-out divided by the wing chord shall be subtracted
from the extreme span as well as the other quantities mentioned in paragraph 3. In
this case the net beam load over the section between cabane struts will not be the normal

Pies

net beam load, wy, but will be equal to 2" wy, where J is the distance between the cabane
2

struts and l, the area of the cut-out divided by the chord. When J, is zero on account of the
center section struts meeting at the center of the upper wing, as is the case with some
designs, no correction will be made in the curve of net beam load, though the correction
to the effective length will be applied. Changes in wing section at gas tanks and similar
design features will be treated in a similar manner.

5. Effect: of different chord lengths.—When an airplane has wings of different
chord length, the modifications to the method of paragraph 3 are as follows:
(a) Two values must be computed for the dead weight per inch run wg. These values
can be obtained from the formulas.

1
Waa ea and Vie Wua
u

ae L,10

Where W,=dead weight of wings in pounds. .
Wya= dead weight of upper wing in pounds per inch run.
W,q= dead weight of lower wing in pounds per inch run.
C,= chord of upper wing.
(,=chord of lower wing.
L,=effectiye span of upper wing in inches.
L,=effective span of lower wing in inches.
If the weights per square foot of the two wings are known to vary, this formula may
itably modified.
. ses computing the relative efficiencies of the wings the gap-chord ratio will be the

ratio of the gap to the arithmetic mean of Cy and C;.

ing the coefficie 5 oe shall b
(c) In computing the normal gross beam load on each wing the coetiicien Gj, Shall be

inserted in the formulas of paragraph 3 (b) V changing them to

Wig pps and Wag= 77 Wie

Gelu tly

(d) The normal net beam loads are obtained by subtracting Wua and Wa from Wy,
and w,, respectively.

6. Effects of decalage or use of different airfoils for upper and lower wings.—
If the wing chords are set at an angle to each other (decalage) or the upper and lower wings
are of different airfoil section, the relative efficiency can not be obtained from Figure 2,
but special wind-tunnel tests should be made to determine the value to be used.

” Effects of ailerons.—lIn the design of the wing spars the load on an aileron used
for computing the stresses on the wing structures shall be assumed the same as on a similarly
located part of the wing—that is, no account need be taken of possible special loads due
to the ailerons functioning as control surfaces except in the design of the aileron and
adjacent portions of the wing structure. Where the aileron projects beyond the tip of the
wing proper, the following modifications shall be made in computing the loads on the
wings:

(a) In computing the design loads on the front truss the projecting portion of the
aileron may be neglected.

(b) In computing the design loads on the rear truss the projecting portion of the
aileron may be neglected, except in computing the bending moments and shears in the rear
spar outside of the outer point of inflection of the outer bay. In that section of the reat
spar the load on the projecting portion of the aileron shall be allowed for as follows:

I. The load on the projecting portion, Wa, will be found by. multiplying the area
of that portion, A, B, C, D, Figure 6-A, by 0.8 times the normal net load per square
foot on the wing.

II. The design shear on the section of the rear spar between the outer strut and
the wing tip shall be that determined by the rules of paragraph 3 plus Wa. The
shear inside of the outer strut shall be computed as outlined in paragraph 3.

III.: The bending moment shall be increased by the amount shown by the curve
G, E, F, of Figure 6-B.

IV. In computing these shears and bending moments, the arbitrary increasé
called for in paragraph 3 shall be applied only to that part of the quantity due to the
load on the portion of the structure inside of the tip of the wing proper. It will not
be applied to the part due to the load on the projecting portion of the aileron.

8. Effect of taper in plan only.—In determining the loads on a wing tapered il
plan only the procedure is similar to that for constant chord wings except that areas ate
dealt with instead of lengths or spans. After the distribution per square foot of area has
been determined the results are multiplied by the appropriate chord lengths to find the11

values of load per inch of span. The detailed modifications in the procedure of paragraph 3
are as follows:

(a) An effective area is computed in place of the effective span of paragraph 3. This
effective area is the area inboard of the effective tip, the location of which is determined
by the methods indicated in Figures 3, 4, and 5. For externally braced wings the section
DE, Figure 3, is located at the outer struts, the tip length L, is represented by BF, Figure
3, and the effective tip is at 0.25 L, from the wing tip. For internally braced wings the
section DE, Figure 3, and B, Figure 5, are located so that the chord equals the distance
to the extreme tip—that is, so that BF equals DE in Figure 3. The tip length, Z,, is repre-
sented by BF, Figure 3, and the effective tip is at 0.10 LZ, from the wing tip.

(b) The dead weight of the wings is computed in pounds per square foot by dividing

d

the total dead weight, Wa, by the total area of wing, A—that is, we= ae

(c) In computing the relative efficiency, the gap-chord ratio and stagger shall be
determined by the dimensions of the intersection of the airfoil sections with a plane parallel
to the plane of symmetry and passing through the geometrical centroid of the effective
area of the smaller wing.

(d) The formulas for the normal gross beam load are modified as follows:

g
Wie = and Wue = CWie
1g eAy = A, ss 1

Where A, and <A, are the effective areas in square feet of the upper and lower wings,
respectively, and w,, and Wy, are expressed in pounds per square foot.
| (e) The normal chord load and the normal net beam load are computed as for a
constant chord wing, except that quantities are in pounds per square foot instead of pounds
per inch run.

(f) The ordinates to the curves of load distribution are found as designated in para-
graph 3. Figures 4 and 5 may be considered as applying to the tapered wing except that
the values are in pounds per square foot. Additional curves should then be drawn to show
the distribution in pounds per inch run. These curves are found by multiplying each
ordinate of the preceding curves by the chord in inches and dividing by 144. If the lead-
ing and trailing edges are straight lines, the curves of net beam Joading over the tip sec-
tion will be slightly curved. It is permissible in this case to compute the ordinates of load
per inch run at each end of the tip section and assume the included portion of the loading
curves to be straight lines.

9. Effect of taper in thickness ratio.—When the thickness ratio of a wing varies,
it is assumed that the air forces and the deadweight of a wing per square foot vary in pro-
portion. The loading curves for such a wing are determined by computing the dimen-
sions of an ‘equivalent wing’ assumed to be of constant thickness ratio and finding the
loading curves for the equivalent wing by the rules given above. The span of the equiva-
lent wing is assumed to be the same as that of the true wing, but the chord at any section
is the chord of the true wing at the section multiplied by the thickness ratio at the section
and divided by the thickness ratio at the section of maximum chord. If the true wing has
parallel leading and trailing edges and the thickness ratio has a straight line variation (the
usual case), the chord of the equivalent wing will also have a straight line variation.

In computing the relative efficiency, the gap-chord ratio and the stagger shall be de-
termined by the dimensions of the intersections of the actual wings (not the equivalent
wings) with a vertical plane parallel to the plane of symmetry and passing through the
geometrical centroid of the effective area of the smaller actual wing.

In certain cases, such as those of metal or ply wood covered wings, where the weight
of the covering forms a large percentage of the total weight of the wings, the department
‘may require that the dead weight of the wings be assumed to be distributed uniformly over
the actual area of the wing instead of being varied in proportion to the thickness ratio.12

  

10. If the true wing is tapered in plan or the thickness ratio has not a straight ling
variation, the chord of the equivalent wing will not have a straight line variation. Th such
a case the equivalent wing should be divided into parts, in each of which a straight line
variation of the chord and loading curves may be assumed. On an externally braced wing
the dividing lines between these parts should be at the interplane strut location. On inter-
nally braced wings these dividing lines shall be at the sections where the chord of the actual
wing equals the distance to the wing tip, the section where the wing is connected to the
fuselage, and a reasonable number of intermediate sections. ‘Bue

11. Effect of several variables.—Where the design of a wing incorporates more than
one of the features discussed in paragraphs 4 to 10 the computation of the loads shall
include all the modifications needed to take care of the special case. Where a designer is
uncertain as to the method of applying these rules to an airplane being designed by him,
he should request the department for a special ruling. The request should be accompanied
by drawings showing the relative location of the wings, their dimensions, interplane bracing,
data as to the airfoil section used, the total weight of the airplane, the dead weight of the
wings, and any other pertinent data that may be available.

12. Application of beam loads to spars.—With two-spar construction the beam
load on each element of the wing in the high and low incidence and inverted flight condi-
tions shall be assumed to act at the center of pressure, and be divided between the spars
in inverse proportion to the distances from the center of pressure location to the spars,
For wings having more than two spars the load shall be distributed among the spars by a
method approved by the department. Table 2 gives the center of pressure location. to be
used for several airfoil sections and is based on its most forward location for high incidence
and inverted flight and on its location when

ky=0.25 ky max.
for low incidence. These values shall be used for all speed ranges.

TABLE 2.—Center of pressure location and ratio of chord to beam components for representative airfoils

 

 

   

|
Center of |

pressure | Ratio

| location in of chord
ieee per cent of ce component

Airfoil chord high} j,¢idence | to beam
incidence ~~~ | component

and in high

inverted incidence

flight

ROEOINGIINOG Qa a ass ak eae ee oe Mere esse ee eaesen oat Pee 29 48 —0. 178
ATOthOGRs 3555 5. Noe een y sabi soa a iad 3. fe = eh gS 3 ince : 30 52 —. 159
Clarke Yt ae fps Se SE ane OE ee oe Dee eae Abb coawee a ak Cpa oga ws 31 51 —. 14
WMG O60 6 rs ee ee Pe ie cea eee 5 a ee See eee eh fad hg 5 fis Se aE Lake 30 51 —, 105
CROUULAP CI 38/2232 eae ee ee Sees ae eee ce see eee ea te : eco upae a 31 52 —. 156
Gottincon tnt.) Sex sees ee. eet: pees sg gie eared eich bb oehe we dee 31 50 —, 135
IN ae Om a aij sant gu ee sah pe EES Tete foe ae Pe ee sete ota ee ; aa Scenes 125 240 —, 225
ReAs Pat6 oe: 29 38 —.127
Sloane 105- -- 30 48 —. 078
U.S. A. 35 B é cee 30 49 —. lil
DB Ne 0 eo ieee vee ac aawec toe ese weee es 27 38 —. 191

 

 

 

 

1 Based on C. P. at ky max.
2 Based on C. P. with ailerons or flaps set at an angle of 25°.

ne ; oars, ed a
The center of pressure location for ‘stable’ airfoils in the low incidence condition

should be obtained from wind tunnel tests made with ailerons or flaps set at 20° or 25°. If

such data are not available, a location between 45 and 50 per cent of the chord should be
assumed.

The ratio of net chord component to net beam component will be assumed as 0.15 in
low incidence and zero in inverted flight for all airfoils. In high incidence, it is obtained
from the formula: C/B= tan (cotL/D—7)

Where C/B =the ratio of the chord component to the beam component.
1= angle of attack when the center of pressure is at its most forward location.13

The lift and beam components of the air load are assumed to be equal as the error
involved is negligible.

13. Application of chord loads to internal wing truss.—The beam loads are to
be assumed to act on the spars as distributed loads although, in practice they are applied
as small concentrated loads at the ribs. The chord loads shall be considered as concen-
trated loads at the panel points of the internal wing trusses. The load at each such joint
shall be the product of the normal chord load per inch run at the joint and the distance
between midpoints of the adjacent panels of the internal truss. For this purpose, the
extreme wing tip shall be considered the midpoint of a truss panel.

14. The nose-dive condition.—In the nose-dive condition the beam loads on the
front spars shall be the same as the design ultimate loads on those spars in the inverted
flight condition. By taking moments about the rear spars it is possible to determine
the load required on the tail surfaces, assumed to be applied at the tail post, to maintain
equilibrium. Since the sum of the forces parallel to the beam direction must be zero, if
the airplane is to be in equilibrium, the total upward load on the rear spars will equal
the sum of the downward loads on the front spars and tail surfaces. This load is distributed
in the usual manner. The chord compdnents of the air load on the wings shall be pro-
portional to the chord loads in low incidence but their total magnitude shall equal the
net weight of the airplane. These chord loads may be assumed to be held in equilibrium
by loads acting on the fuselage structure at panel points near the center of gravity. The
loads on the wing framework shall be assumed to be carried by the incidence bracing,
interplane struts, rear flying wires and drag truss members only, unless a least-work analysis
is made. For internally braced cantilever monoplanes the loads specified above shall be
increased 30 per cent. This nose-dive condition shall be investigated regardless of the
airfoil used or the type of airplane being designed.

15. Dead loads carried by the wings.—Where dead weights are carried by the
wings, such as landing lights, wing-tip floats or skids, gas tanks, etc., their effect shall be
allowed for as follows: If the weight is small and in such a location that it has little
effect on the computed stresses, it,may be included in Wa, the total dead weight of the
wing. Loads that are large or are so placed that they would cause an appreciable change
in the computed stress system shall be separately provided for. If the load is the weight
and inertia of a compact unit, it shall be treated as a concentrated load on each spar,
the relation between these loads depending on the location of the center of gravity of the
unit. If the unit is spread over a considerable extent of wing area, like a gas tank or a
wing radiator, it shall be considered either as composed of concentrated loads at the points
where it is supported by the spars, or as_a uniformly distributed load, depending on the
manner of support. If a dead weight is supported in such a manner that it can be con-
sidered as a uniform load and extends over two-thirds or more of a panel of the lift truss,
it may be assumed as distributed over the entire panel. When the unit is one of variable
weight, as a fuel tank, the primary analysis shall be made, using the maximum weight.
An investigation using the minimum weight shall also be started and carried far enough
to determine the design of any member limited by this condition or to show clearly that
there are no such members. ‘The weights of all such loads shall be assumed. to act in the
beam direction in the high and low incidence and inverted flight conditions, and in the
chord direction in the nose dive. The effect of nacelles of multi-engine airplanes and
their contents shall be computed according to the same basic principles, the application
of which to the special case is outlined more fully in paragraphs 41 and 42.

16. Internally braced biplanes shall be designed by a method acceptable to the
department; that described in Air Corps Information Circular (Aviation) No. 440 is
recommended.

17. Loads on control surfaces.—The average ultimate loads to be used for the
design of control surfaces are given in Table 1. The loads on the fixed surfaces are to
be assumed uniformly distributed, but on movable surfaces the intensity of loading at
the hinges shall be equal to the loading on the fixed surfaces in front of-them, and shall14

decrease uniformly to an intensity at the trailing edge of one-third the former value,
Portions of the movable surface in front of the hinge line, except auxiliary vanes, shall
carry the same loading as the fixed surface.

(a) When auxiliary vanes are used for balancing, they shall be assumed to be sub-
jected to the same intensity of loading as the fixed surfaces, when computing the
distribution of load and the stresses in the remainder of the movable surface. The vanes
themselves and their attachment to the main movable surface shall be strong enough to
carry the load required to balance the load on the main portion of the movable surface,

(6) The control surfaces must be designed to carry the specified load acting in either
direction.

(c) Although no load paraillel to the chord of the fixed tail surfaces, or torsional load
is specified, provisions shall be made to carry a reasonable amount of such load.

18. Loads from control system.—The forces in the control wires or push rods oper-
ating the movable surfaces must be computed and their effect on the rest of the structure
investigated and allowed for in the design of the structure. Special care must be taken
to make sure that the portion of the wing between the rear spar and the aileron is strong
enough to carry the load for which the aileron is designed, and the structural members in
this part of the wing shall be analyzed in the aileron analysis.

19. Landing gears—Level landing.—The resultant of the ground reaction is
assumed to be a force lying at the intersection of the plane of symmetry, and a plane in
which are located the-axles and the center of gravity of the airplane. The propeller axis
is assumed horizontal, and the basic value of the vertical component of the resultant of the
ground reaction equal to the gross weight of the airplane minus chassis and wheels. The
horizontal component is of the magnitude required to give the resultant force the
specified direction. The resultant of the ground reaction is assumed to be divided equally
between wheels and to be applied at the axles. The shock absorbers shall be assumed to
be fully extended unless a partial extension of not less than one-half the total extension
would cause greater stresses, in which case such partial extension will be assumed.

20. Landing with side load.—The landing gear and cabane and that part of the
fuselage adjacent to them should be analyzed under the following combination of loads:

(a) The loads specified in paragraph 19 for the level landing condition.

(b) Side components of the ground reactions equal to one-fourth of their vertical
components. ‘These side components of the ground reactions shall be assumed to be
equally divided between wheels and to act at the ground, the tire being assumed to have
defected one-half the nominal diameter of its ctoss section.

(c) Side components of the weight and inertia forces equal to one-fourth of their

vertical components. These side components may be assumed to act on the fuselage |

  

 

 

 

 

 

jomts to which the wings and chassis are attached, and to be divided between joints in |
such a manner that their resultant would pass through the center of gravity of the airplane. |

(¢) A couple composed of vertical forces acting at the main connections of the wings
to the fuselage and cabane and of sufficient magnitude to bring the entire system into
equilibrium.

The required load factor will be one-half that specified in landing conditions, Table 1.

In the case of the formerly conventional V type of chassis, the spreader tubes and
diagonal brace wires may be designed to carry a load of 1.5 times the weight of the air-
plane acting in either direction and applied to one wheel at its axle, in place of the loading
specified above.

#1. Three-point landing.—The basic value of the sum of the ground reactions is
the gross weight of the airplane. The total load is divided between the chassis and tail
skid in inverse proportion to the distances, measured parallel to the ground line, from the
center of gravity of the airplane to the points of contact with the ground. The net basic
load on the chassis shall be equal to the total load on the chassis minus the chassis weight;
and it shall be divided equally between wheels. Loads are assumed to be perpendicular15

‘to the ground line in the three-point landing attitude with all shock absorbers extended to
‘the same degree as in the level landing condition.

22. Reactions.—In all loading conditions for the landing gear, the reactions of the
fuselage or wings are the forces required to balance the net loads on the axles. The
equilibrants of these reactions will be used in the analysis of the fuselage in the landing
‘conditions.

23. Braked landing.—Airplanes equipped with brakes shall be investigated for the
loads incurred when a landing is made with the wheels locked and the airplane is in an
attitade such that the tail skid just clears the ground. The full weight of the airplane
will be assumed to act on the wheels in a direction perpendicular to the ground line in this
attitude. In addition a component parallel to the ground line shall be assumed to act at
the point of contact of the wheel and the ground, the magnitude of this component being
equal to the weight of the airplane times a coefficient of friction of 0.55, or equal to the
load required to keep the airplane in equilibrium with the tail skid just clear of the ground,
whichever is the less. The chassis structure shall have sufficient strength to withstand
the axial loads, bending, and torsion to which it will be subjected under this combination of
loads, the load factor being one-half that specified in Table 1 for the landing conditions, with
no reduction permissible because of the use of oleo shock absorbers.

24. Seaplane landing with inclined reactions.—Assume propeller axis horizontal
and the resultant water reaction acting in the plane of symmetry and passing through
the center of gravity of the airplane but inclined so that its horizontal component is equal
to one quarter of its vertical component. The forces representing the weights of and in
‘the airplane act in a direction parallel to the water reaction. The weight of the floats and
float bracing should be deducted from the gross weight of the airplane. A load factor of 8
lis required.
| 25. Seaplane landing with vertical reactions.—Assume propeller axis horizontal
and the resultant water reaction vertical and passing through the center of gravity. The
weight of the floats and float bracing should be deducted from the gross weight of the
airplane. A load factor of 8 is required.
| 26. Seaplane landing with side load.—<Assume the airplane level transversely and

the propeller axis horizontal; neglect the vertical forces and assume a side force equal to
the gross weight of the airplane, applied to the side of the main float or, in the case of
twin float machines, distributed equally between the floats. The side force may be con-
sidered as a concentrated load applied to the side of the float at a point m the transverse
vertical plane containing the center of gravity of the airplane and approximately halfway
between the bottom of the keel and the level of the water line at rest. A load factor of
2 is required.

27. Fuselages.—The fuselage shall be investigated for seven loading conditions;
high incidence, low incidence, maximum stabilizer and elevator load, maximum fin and
rudder load, level landing, three-point landing and nosing over. The entire fuselage
‘structure will not have to be analyzed for each of these conditions but all must be con-
sidered and each stress investigation carried far enough to insure that the loads in all
‘members limited by that condition are computed. Certain additional stress investigations
will also be needed where eccentricities or bending loads occur and the results used in
‘connection with the standard loading conditions in the final design of the member affected.
By careful planning it will be possible to obtain the maximum.loads in all members with
‘a minimum of work and in the description of the various loading conditions, this phase
of the problem will be considered. Usually the analysis of the fuselage should not be
‘made until after those of wings and landing gear. When it is not desired to carry out
the work in this order, the fuselage analysis should be corrected in accordance with the
results of the other analyses,

| 61109278rape

16

28. Table of panel loads.—

 

 

TABLE 3
————————— —<—<———————— — SS Reet ——— fr i ——_. 58
| ; Vertical Vertical | Horizontal | Panel points 1, 2, 3, 4, | Horizontal
Unit Weight arm moment | arm 6tCss 8; DB; ¢; d; 2 | moment

|
|
Wngine 2... 22-0 Ss. sano ae nea eae
|

Center of gravity. -----------------------------------0 20 |
| |
Bataleon Se ee enews eer | | | |

In the horizontal line beginning “Horizontal arm’’ fill in the distance from the datum
point to the panel points in question in the spaces occupied in the sample table by a,
b, ¢, ete.

In the vertical column under “Horizontal arm” fill in the horizontal distance from
the datum point to the center of gravity of the unit considered.

In the column headed ‘Vertical arm”’ fill in the corresponding vertical distance.

Any point may be used as the datum, but the point selected should be clearly stated.
The front face of the rear propeller flange will usually be found convenient and is
recommended.

In the column under the heading “Panel points” should be inserted the portion of
the weight of each unit carried at each point. The horizontal totals will be the weights
of the units, and the vertical totals, the panel loads.

All weights should be given including wings and chassis, but the panel loads should
be modified when certain weights are omitted in the fuselage analysis, as, for example,
the weight of the wings in the high incidence condition.

Panel points may be numbered either front to rear, or vice versa.

As many horizontal and vertical lines should be used as are necessary to represent
the design in question, the form shown being only a guide.

In computing the panel point loads to be tabulated as indicated in Table 3, the following
rules shall be observed:

(a) Items located between two adjacent panel points of the side trusses shall be’
divided. between those panel points in inverse proportions to the distances from them to
the center of gravity of the item.

(b) Items to the rear of the tail post shall be represented in the table by a load anda
moment at the tail post.

(c) Items supported at three or more panel points shall be divided between those
points by the aid of an investigation and analysis of the method of support, if practicable.
Where a rational analysis is not possible the division may be estimated.

(d) In all cases the moment of the partial panel loads due to any item about an origin
near the nose of the fuselage shall be equal to the moment of the item about that origin.

(e) If desired, the loads may be divided between the upper and lower panel points
in proportions determined by the vertical location of the units they represent. All loads
are assumed to lie in the plane of symmetry and to be divided equally between the two
vertical trusses of the fuselage.

29. Panel point loads in three-point landing.—A separate table of panel point
loads should be computed for the three-point landing condition unless the loads are care-
fully divided between upper and lower longerons, as suggested in the preceding paragraph.
The labor involved can be greatly decreased by grouping items that are supported by the
same panel points and treating the group as a unit. The distance of the centroid of the
group from the origin measured parallel to the ground can be found from the formula:

, uly cos 9+ 2M, sin 6

=W17

Where x’ = distance from the origin to centroid, measured parallel to the ground line.
2 M,=sum of horizontal moments of units in group from balance schedule.
> M,=sum of vertical moments of units in group from balance schedule.
>W=sum of weights of units in group from balance schedule.
6=angle between ground line and propeller axis.
The distance measured parallel to the ground line from the origin to the panel points
ean be scaled from the drawing or computed from the usual formula for the rotation of axes.

xz’ =x cos 0+y sin 0

Where x’ = distance measured parallel to ground line.
x = distance measured parallel to propeller axis.
y = distance measured normal to propeller axis.
6=angle between propeller axis and ground line.
y is positive if the point is above the propeller axis and negative if it is below.

Whichever method is used to determine the panel loads for the three point landing
condition, care must be taken that the equations of equilibrium are satisfied by the panel
loads and reactions used.

30. Low incidence condition.—In the low incidence condition the fuselage shall
be assumed to be subjected to the following loads:

(a) Weight of fuselage and contents multiplied by the low incidence load factor.
These weights are assumed to act normal to the propeller axis.

(b) A load at the tail post, or on the horizontal tail surfaces perpendicular to the
propeller axis, and of the magnitude required to obtain equilibrium.

(c) Loads from the wings equal to the forces supporting the wings in the low incidence
condition multiplied by a factor sufficient to make the sum of the forces perpendicular to
the propeller axis equal to zero.

(d) A load along the propeller axis sufficient to make the sum of the forces parallel
to that axis equal to zero. The loads and factors under (6), (c), and (d) are to be found
by: applying the equations of equilibrium to the entire structure. The entire fuselage
side truss and some crossties will usually have to be analyzed for this loading condition.

31. Three-point landing.—In the three-point landing condition the fuselage shall
be assumed to be subjected to the following loads:

(a) Weight of the entire airplane excepting only such units as are carried by the
landing gear without the aid of the fuselage. These loads are assumed to act perpen-
dicular to the ground line. The load factor for this condition is specified in Table 1.

(b) The landing gear reaction, which, as determined from the landing gear analysis
for this condition, should balance the weights. The entire fuselage side truss will usually
have to be analyzed for this condition.

32. Maximum stabilizer and elevator load.—The fuselage is to be analyzed under
the maximum loads on the horizontal tail surfaces specified in Table 1 and the required
supporting forces at the wing connections. The load on the tail is assumed to act down.
Usually it will be found that at some section to the rear of the connection to the wings, the
loads in the fuselage members in this condition are less than in low incidence. The analysis
for this loading need not be carried forward of that-section. In designs without tail skids,
as seaplanes, a similar loading condition must be investigated in which a load equal to half
that specified in Table 1 is assumed to act upward on the stabilizer and elevator.

33. Maximum fin and rudder load.—The fuselage is to be analyzed under the
maximum loads on the vertical tail surfaces, specified in Table 1 and the required supporting
forces at the wing connections. These loads must be assumed to act in either direction.
In analyzing for this condition proper allowance must be made for the torsion due to the
fact that the center of pressure on the vertical surfaces is above the fuselage, where that is:
the case. On many airplanes, this can best be done by applying a load greater than the tail
load on the upper truss and a load in the opposite direction on the lower, the resultants of_ structure in a manner similar to that suggested in paragraph 30 for computing the loads m

The load factor required shall be that for the three-point landing condition.

-. Where T'= torque in inch pounds

 

18

t load on the tail. On small airplanes even this
conservative method will indicate design sizes smaller than could be allowed from other
practical considerations.. Where this method is considered too conservative, the method of
least work or some similar method of allowing for the torsion may be used. In all cases,
proper account shall be taken of cockpit openings and similar places where the trussing ig
not continuous. This condition will usually determine the design of the top and bottom
trusses of the fuselage except in the forward portion where they are affected by the wing and

these two loads being equal to the resultan

landing gear loads. e
34. Level landing.—In the level landing condition the fuselage shall be assumed t9

be subjected to the following loads: ;
(a) Weight of the airplane, except the chassis, assumed to act normal to the propeller

xis.
oS (b) The landing gear reactions from the level landing chassis analysis. The vertical
component of these reactions should balance the weights. |

(c) Horizontal loads representing the inertia forces due to slowing down of the airplane,
As the rear portion of the fuselage will not be limited by this loading condition, the following
procedure will be found convenient.

(d) Compute the loads in some panel to the rear of the wing connections by the method
‘of sections or its equivalent, and apply these loads as external forces at the forward end of the
panel. In making this computation consider only the weights of fuselage and contents to
the rear of the section investigated.

(e) Divide the load under (c), into a load on each longeron at the forward end of the
panel mentioned in (d), in such manner that the front part of the fuselage truss will be in
equilibrium under the external forces. This condition will usually be critical only for
members in the forward part of the fuselage. If desired the panel loads may be assumed
to act parallel to the resultant of the landing gear reactions.

35. High incidence.—In the high incidence condition, the fuselage shall be assumed
to be subjected to the following loads:

(a) Weight of fuselage and contents multiplied by the high incidence load factor,
These forces are assumed to act normal to the propeller axis.

(b) The high incidence wing reactions multiplied by the factor required for equilibrium,

(c) A load on the nose of the fuselage, acting along the propeller axis, and of the
magnitude required by the conditions of equilibrium. ‘This load may be considered to
represent the resultant of the propeller thrust, drag of the fuselage, and the components of
the weight and inertia forces of the fuselage and contents parallel to the propeller axis.

(d) An arbitrary load at the tail post, acting normal to the propeller axis, of the magni
tude, and acting in the directon required by the conditions of equilibrium.

(e) The torque load treated in the manner described in paragraph 37. In order to
determine the factor required for load (6), and the magnitude of loads (c) and (d), it will
usually be necessary to write down and solve the equations of equilibrium for the entire

low incidence.

36. Nosing over.—The front part of the fuselage shall be designed to resist the forces
to which it would be subjected in nosing over unless the landing gear is of such construction
that the probability of such an accident is remote. To simulate this condition, assume the)
airplane to be resting on the wheels and the center of the propeller hub or that portion of the
primary structure of the fuselage that would strike the ground first. Assume the gross
weight of the airplane to act at the center of gravity and perpendicular to the. ground.

37. Computation of torque load.—The engine torque is equal to:
T= 63,000 P/n

P=horsepower of engine
m=engine speed inr. p.m. (For geared engines n= propeller speed.)iy

The torque load on each engine bearer is 7/d where d is the distance in inches between
engine bearers. This load acts down on one engine bearer and up on the other.

The resulting moment is taken care of by an unsymmetrical distribution of load
between the wings and by forces in the fuselage cross bracing. In certain cases especially
when geared engines are used, the stresses due to the torque should be computed for all
fuselage members affected, the necessary reactions being assumed to act at the connections
of the wings with the fuselage. Otherwise, the following approximation may be used.
The torque load is assumed to act down on the engine bearer and to be held in equilibrium
by vertical forces acting at the main connections of the wings with the fuselage, the engine
bearer and the members of the fuselage side truss being assumed to lie in a single plane
parallel to the plane of symmetry. In case of direct-drive engines, the stresses due to
torque load and its equilibrants shall be multiplied by a factor of safety of 2 and added to
the stresses found for the high incidence condition; in the case of geared engines, the
factor of safety to be used shall be 3. The stresses are to be added arithmetically and not
algebraically, because if the forces induced by the torque load in any member are opposite
in sign to those due to the dead weights, there will normally be a corresponding member
on the opposite side of the fuselage in which the forces due to torque loads and dead
weights will be of the same sign.

When a direct-drive engine is carried by engine bearers that are supported at two
or more points, the torque’load shall be divided between the points of support in the same
proportions as the weights carried by the engine bearer. When a direct-drive engine is
supported by a vertical plate, the torque shall be assumed to act at the center of gravity
of the engine. In the case of a geared engine, the torque shall be assumed to act at the
gear box.

88. Special conditions.—In addition to the standard loadings described above,
which give only the axial loads in the side trusses the following points must be investi-
gated and computations made, if necessary.

(a2) If units situated between panel points or loads from other major assemblies
are carried by bending in members of the main trussing, numerical estimates of the amount
of such bending must be made and, if serious, this bending must be allowed for in the
design of the affected members.

(6) The strength of the fuselage members, not part of the main trussing, but trans-
mitting loads to it, must be investigated. In general, these members shall be designed
for the same load factor as the part of the main fuselage structure to which they are
connected.

(c) Care should be taken that the fuselage is strong enough to carry the loads from
the chassis in the level landing condition with side load, or in a one-wheel landing condition.
The loads from the chassis in these cases are unsymmetrical and the bulkheads where the
fuselage and chassis are connected should be strong enough to equalize them between
the fuselage trusses.

(d) The engine mount section of the fuselage shall be capable of carrying the weight
of its contents, acting sideways, with an ultimate load factor equal to one-quarter of the
high incidence factor.

(ec) Wherever there is an eccentricity in the fuselage trussing, the stress diagram
shall be drawn for a theoretical truss, in which the eccentricity has been eliminated. In
designing the members, however, the bending due to the eccentricity shall be computed
and used in determining the design sizes.

(f) The forward portion of the fuselage shall be analyzed for the condition of inverted
flight in all cases where it is not necessary to analyze this part of the structure for the
nosing-over condition.

(g) Design of new engine mount for existing airplane.—When a new engine mount is
to be designed for an existing airplane and it is either undesirable or impracticable to
make an entire new fuselage analysis, it shall be designed for the following conditions:|

20 |

I. High incidence and torque: The analysis shall be made as specified for the)
|
}

fuselage for those conditions, except that the reactions
be applied at the points where the new structure 1s joined to the structure of the)

required may be assumed {9

original design. q
If. Inverted flight: The load in each member shall be assumed to be of opposite:

zs 10° 1 >] IA ¢ : "0 Be 36 ita 90N)
sign to the load in that member, due to high incidence and torque, and its magnitudgl
equal to 75 per cent of that load. This condition will be assumed to cover the coms
dition of nosing over as well as inverted flight, the high ratio of reversal of load being)
used so this will be possible. 8 . .
III. Side load: This condition shall be analyzed as specified above in subs

3

paragraph (qd). — q

IV. In designs for which the design load factor for the landing conditions ig)
ereater than that for high incidence, the landing factor shall be used in place of ther
high incidence factor in the analysis for the condition of high incidence and torqued

(h) When the point of contact of the tail skid with the ground is in front of the tail post a
distance greater than 1.5 times the vertical distance between longerons at the tail skidj
station, the rear portion of the fuselage shall be investigated for a special three-point}
landing condition in which the basic load on the tail skid shall be assumed to act at the,
lower end of the tail post. The load factor for this condition shall be that for the landing)
conditions minus 1. |

39. Cabane.—lIn the inverted flight and nose-dive conditions, the cabane should be
considered a part of the wing structure and analyzed with it. In the high and low incideneg
conditions it will be simpler, in general, to analyze it as part of the fuselage structure
When there is some doubt as to how the loads are divided between cabane members, all
possibilities should be considered and the analysis made according to assumptions thag
will be conservative for both cabane and fuselage. This may require more than one analysis
of the cabane and forward section of the fuselage in either high or low incidence. Thé
most usual case where there will be doubt regarding the action of the cabane is when the
front spar is attached to the fuselage by a tripod of struts. In this case the cabane should
be investigated by the method of least work.

In order to allow for unsymmetrical loads on the wings, as in a roll, the cabane shall
be investigated for the stresses that would be present in the high and low incidence cons
ditions if 75 per cent of the design load were acting on one wing and 60 per cent of the
design load on the other.

40. Nacelles.—The loads on a nacelle must be considered from two standpoints—that
of determining the design of the nacelle itself and that of determining the design of the
internacelle wing structure. For the design of the nacelle itself the following conditiong
must be considered:

(a) Weight of nacelle and contents acting in beam direction multiplied by high incr
dence load factor, torque load multiplied by a load factor of 2 and reaction from wing
structure.

(6) The weight of nacelle and contents acting sideways multiplied by one-quarter the
high incidence load factor and reaction from the wing structure.

(c) Condition (a) with the direction of loads reversed and the inverted flight load
factor. The torque load may be decreased proportionately to the other forces. 7

41. Effect of nacelles on beam forces.—In the wing analyses, for high and low
incidence and inverted flight, the weight of the nacelle and contents must be treated as@
beam load on the wing structure. The division of this load between points where thé
nacelle is supported will be obtained from a nacelle analysis. Usually it will be simplest
to make the nacelle analysis in such a way that the effect of the torque and variable weights
can be separated from the effect of the fixed weights. The nacelle load on each trus
can then be considered to act as a single load at the nacelle center line or as equal load
at the sides of the nacelle, depending on the local construction. Two sets of conditions
must be investigated—one with both fixed and variable weights and the other with onl21

the fixed weights considered. The torque may be neglected in figuring the effect of the
nacelle in the wing structure, as its effect is simply to shift the center of gravity of the
nacelle slightly.

42. Effect of nacelles on chord forces.—In designs with nacelles the followin
rules should be used for computing the loads in the internal or drag trusses:
| (a) The internal wing trusses shall be assumed to be supported at the cabane and
fuselage in all flying conditions.
| (b) The head resistance of the nacelles shall be represented, in each of the loading
conditions, by chord loads equal to 0.4 of the chord components of the total air load on
the wing, divided equally betweeen nacelles and acting along the propeller axes. These
loads will be assumed to act backward.

(c) The chord components of the weight plus inertia of the nacelles and contents
shall be computed by multiplying the weight of the nacelles and contents by the load
factor and the ratio of the chord component to the beam component of the air load on the
wings, in the high-incidence and low-incidence conditions, and by the safety factor 2 in the
nose-dive condition. ‘These forces shall be assumed to act at the propeller axes and in the
direction opposite to the chord components of the air load.
| (d) When nacelles are placed on the wings of monoplanes, the same rules will apply,
except that in the nose-dive condition all loads will be increased 30 per cent, as specified
in paragraph 14.
| (e) Rule (6) is based upon the assumption that the total drag of the airplane is pro-
duced 50 per cent by the wings, 30 per cent by the fuselage, and 20 per cent by the nacelles.
Whenever sufficient data exist for a particular design to justify a different assumption as
to the division of the total drag, a special ruling should be requested. Rule (c) is based
upon the assumption that the weight and inertia of the wings act in a direction directly
opposite to the resultant gross air load.
| 43. Effect of nacelles on fuselage analysis.—The presence of nacelles in a design
will affect the analysis of the fuselage principally through changing the location of the
power plant and the amount of the forces from the wings. In order to balance the force
system, it will usually be necessary to apply arbitrary forces at some of the wing connec-
tions, and the amounts of such forces should be definitely stated. In the landing condition
allowance should be made for the way in which the weights of wings, nacelles, etc., are
applied, if at all, to the fuselage structure.

44, Effect of landing gear attached to the wings.—When landing gears are
attached directly to the wings, that part of the wing structure between the outermost
Janding gear members must be considered as a part of the fuselage in the landing conditions.

PART 2. MISCELLANEOUS REQUIREMENTS

1. Wings.—(a) Fittings shall be designed to carry loads 10 per cent in excess of the
design loads for the members to which they are connected. Since, in small wires, the
initial tensions due to rigging may impose a considerable load on the fittings, it is required
that wires and cables up to and including the 3,400-pound size shall be attached to lugs
having the same rated strength as the wires or cables. (See fig. 9 for sizes of standard lugs.)
Heat treated or alloy steel fittings should be used with caution because of the elementof
danger involved in making field repairs.

(b) When streamline wires are used for lift wires they shall be double unless the wings
are so designed that with any lift wire removed the strength of the remaining wing structure
will be adequate to develop load factors of not less than 50 per cent of those specified in
Table 1. .

(c) The maximum wing rib spacing permitted with fabric covering is given in Table 4.
This spacing is determined by the necessity of preventing the covering from deflecting to
‘such an extent that the properties of the wing will be affected by the change in the effective
‘airfoil section. Usually at least one intermediate false rib will be required in the nose

portion to keep the surface free from irregularities.22

Tape 4.—Maximum rib spacing im inches

aS SS a ad

| In slipstream Outside of

   

High speed in miles per hour slipstream
Sees BP gas ee ee | 4
a ae er --<------ a rs 20
1001 eR BE ee ee 3 id 0
Dd paades 30 oe petendiig a i a eda Ss : "a
(AUPE RS See eek oes SA eee estes ie eee er , M2
SN a a ee 8 15

FOO ee ee ee ee eee Sana eer ee ‘ | |

1 For speeds less than 100 miles per hour, use the figures for 100 miles per hour.

(d) Thin ply wood or metal covering should be used on both surfaces of each wing
within the slipstream, between the leading edge and front spar. On all airplanes whose high
speed is greater than 100 miles per hour it is advisable to extend this covering along the
entire leading edge.

(e) In the design of wing spars and other members subjected to combined axial and
transverse loads proper allowance must be made for secondary stresses. 1 he Be rry method
eiven in Aeroplane Structures by Pippard and Pritchard, the method given in Graphische
Ci adil by Muller 3reslau, or the Army Air Corps ‘‘precise”’ method, given in Airplane
Design by A. 5. Niles and in Information Circular 493, are acceptable. Of the three, the
last is preferred. .

(f) In determining the allowable stresses in members subjected to combined bending
and compression the allowable stress will be assumed to lie between that which the member
would carry as a column and that which it would carry asa beam. The exact value will be
determined from the ratio of stress due to bending to the total stress; that is, if the stress
due to bending is three-fourths of the total stress the allowable stress under combined load-
ing will be equal to the stress that the member would carry as a column plus three-fourths
of the difference between its modulus of rupture in bending and its strength as a column,
The allowable stress in wooden members shall be determined from figure 7.

(g) Tests on I and box spars have shown that a decrease in the web thickness results
in a serious decrease in the modulus of rupture. For purposes of design, the modulus of
rupture of such spars shall be that obtained from figure 7. In addition to decreasing the
modulus of rupture, thin webs result in increased deflections and secondary stresses. The
study of the problem has not yet progressed to a point where it is possible to state the most
efficient web thickness for a given spar. In the case of solid spars the web should be at
least one-half inch thick except on very small airplanes, and it will often be advantageous
to use a solid rectangular section instead of anI. Less is known about the proper web
thickness for box spars, but one-eighth inch is considered a good minimum thickness for
sach web. The web may be made of either two-ply or three-ply material, but in either
case the grain of the individual plies should make an angle of 45 degrees with the axis of
the spar. Where box spars are used it is desirable to check the proportions of the design
by constructing a short section of spar and testing it to the required shear and bending
moment.

The allowable total unit stress in spruce members subjected to combined bending
and compression shall be computed by the method developed by the Forest Products
Laboratory, through the aid of Figure 7. On the right-hand side of this figure are two
families of curves, the upper one for the determination of the modulus of rupture, the
lower one for the elastic limit in bending. Each of these quantities is dependent on the
ratio of compression flange thickness to total depth of beam and the ratio of web thickness
to total width of beam. On the left-hand side of the figure are two additional families of
curves. The ‘“horizontal’’ family indicates the elastic limit under combined bending and
compression and the ‘‘vertical’’ family, the effect of various slenderness ratios on the
former quantity.23

The allowable total stress under combined load, F,, is found as follows:

I. For the cross section of the given beam, find the elastic limit in bending and
the modulus of rupture from the ratios of compression flange thickness to total depth,
and web thickness to total width, locating points such as A and B.

II. Project points A and B to the central line, obtaining points such as C and D.

III. Locate a point such as E, indicating the elastic limit of the given section
under combined bending and compression. This point will be at the intersection of
the curve of the “horizontal” family through C and the curve of slenderness ratio
corresponding to the distance between points of inflection.

IV. Draw ED.

V. Locate F on ED, with an abscissa equal to the computed ratio of bending to
total stress. The ordinate of F represents the desired value of Fy.

The following rules should be observed in the use of Figure 7.

I. The slenderness ratio should be that between points of inflection. When the
ends of a member are restrained, this need not be computed with excessive precision,
as a small error in L/p will not result in a large error in F,. For wing spars, the follow-
ing approximations are acceptable:

In computing the margin of safety, L may be taken as the distance between points
of inflection under side load alone, unless the designer wishes to compute this distance
by a precise formula.

In computing the margin of safety near the outer strut, L may be taken as
twice the distance from the support to the outer point of inflection.

In computing p for the purpose of applying the curves of Figure 7, filler blocks
may be neglected and, in case of a tapered spar, the average value should be used.

IT. In computing the modulus of rupture and the elastic limit in bending the prop-
erties of the section being investigated should be used. Filler blocks may be included
in the section for this purpose and also in computing the imposed stresses fp, f,, and f;.

Ill. The bending moment from which f, is computed must include an allowance
for secondary bending. When possible, this should be done by using a precise method
for computing the bending moments. When this can not be done, a rough allowance
must be made for their effect.
For beams subjected to a uniformly distributed lateral load, the distance between

points of inflection can be obtained from the nomogram Figure 8.

(h) At points of application of large concentrated loads, the routing of I spars shall be
omitted, and filler blocks used in the space between the flanges of box spars. These filler
blocks must be designed to carry the local stresses efficiently and to avoid undue con-
sentrations of stress. In order to avoid an abrupt change of section at the ends of the
blocks, they should taper with an angle of not more than 30° to a thickness of not more
than one-eighth inch. The middle portion of the block may, if desirable, have sides
perpendicular to the spar axis. The size and weight of filler blocks can often be reduced
and the tendency to check eliminated by constructing them of ply wood, the individual
Dlies being from one-eighth to one-fourth inch in thickness. The plies adjacent to the
spar webs should have their grain parallel to the spar axis. =
) (1) In computing the area, moment of inertia, etc., of wood spars pierced by bolts,
she diameter of the bolt hole shall be assumed to be 25 per cent greater than the diameter
of the bolt. ;

(7) In computing the properties of box spars for final design, only that portion of the
veb with its grain parallel to the spar axis and one-half of that portion of the web with its
train at an angle of 45° to the spar axis shall be considered. If desired, the more con-
‘ervative method of neglecting the web entirely may be employed. When computing the
orm factor for box spars the total thickness of both webs shall be used.24

(k) For preliminary work, a box spar may be considered as a truss wiih a continuous
web. The total load in each flange, due to bending, will then be the bending moment
divided by the distance between the centroids of the two flanges. vig ne this value by
the flange area will give the average unit stress in the flange. The aretnee unit stresses
in each flange due to bending and axial load should be added algebraically to find thes
total unit stresses. In this method the allowable net stress in the compression flange)
should be taken as 5,000 pounds per square inch and in the tension flange as 10,000 pounds
per square inch, for first class spruce. 7 . | : |

(1) All spruce spars should have a ratio of height to width less than 5 whether they are
of box or solid construction, unless particular attention has been paid to lateral support,
In box spars the area of the tension flange shall always be at least one-half the area of ther

compression flange.

(m) In designing the we
be computed by the formula: f
allowable unit stress may be assumed to be 1
spruce-poplar ply woods with the grain of all
unit stress is used, the webs must be reinforce
their equivalent. If no stiffeners are employed, a lower unit stress 1n shear must be used,
Tests at the Forest Products Laboratory ‘ndicate that a more economical beam design
can be obtained if a lower value of the unit shearing stress is used.

(n) When the deflections of box, solid rectangular or I spars are desired ‘they should
be computed by one of the common formulas of mechanics. Deflections of trusses, no
matter how shallow, should be computed by one of the standard methods applicable to
trusses. ‘The use of a formula applicable to beams for the computation of truss deflections
will not be approved except where tests have been made on the actual truss to determine
its effective “EI.”

(0) Where a joint in a spar is designed to transmit bending from one section of the
spar to another, the stresses in each part of the structure shall be calculated on the assump-
tion that the joint is 100 per cent efficient and also under the assumption that the bending
moment transmitted by the joint is but 50 per cent of that obtained under the assumption)
of perfect continuity. Each part of the structure shall then be designed to carry the more
severe of the loads coming on it as determined from the above assumptions.

(p) If eccentric fittings are used on any member due allowance must be made for the
resulting changes in stress distribution. The effect on the member itself must be carefully
investigated.

2. Fuselage.—(a) In all fuselages, the longerons, vertical, and diagonal struts shall
be designed with a coefficient of restraint between | and 2, according to the judgment of
the designer and the type of fitting. In all cases the value assumed for the coefficient of
restraint must be stated by the designer.

(6) Except for airplanes weighing less than 1,000 pounds the minimum sizes of tubmg
recommended for the primary members of wire-braced fuselage structures are:

34’’ x 0.035 for longerons and web members of the side trusses and members of the
engine mounting.

5%’’ x 0.035 for web members of the top and bottom trusses.

For fuselages of all tubular construction each of these sizes may be reduced one-eighth
inch. |

These sizes are advisable as minima, where their length will not be greater than 30
36 inches, from the standpoint of maintenance and handling, although they may appeaf
to be greatly overstrength for the loads imposed on them in the design conditions. For
larger fuselages the diameters given above should be increased, or the tubes made suffi
ciently strong to carry a load equal to the sum of the vertical, or horizontal, components ol
all wires. attached to the member in question when such wires carry an initial tension 0
25 per cent of their rated strength.

bs of box spars the maximum shear stress intensity should
=SQ/bI, where ‘‘b” ig the entire web thickness. The
,800 pounds per square inch for spruce oF
plies at 45° to the spar axis. When this
ed by internal diaphragms or stiffeners oF25

(c) The attachment of items of equipment, etc., to main members of the fuselage
structure should be made by means of clips or pins, not by welding.

3. Landing gear.—(a) Shock absorber cord installations shall be designed to with-
stand the shocks of landing, and care must be taken in construction that the cord is applied
with the proper initial tension to give the shock absorbing characteristics upon which the
design was based. ;

(6) The effect of eccentricities in the chassis due to the rise of the axle or the relative
locations of other members shall be carefully considered in the design.

4. Fittings.—(a) The dimensions of all lugs should be such that they will fit standard
clevises of corresponding sizes. The average strength of lugs of any design shall be at
least 15 per cent in excess of the rated strength, and the actual strength of any individual
lug must not be below the rated strength. Lugs made of cold rolled steel, in conformity
to the limiting dimensions given in Figure 9 are considered to comply with the above
requirements. Lug designs of other materials, or not in conformity with the dimensions
of Figure 9, will have to be tested before acceptance. Such tests must be carried to destruc-
tion of the lug and the type of failure reported. The following rules govern the use of
Figure 9.

I. Dimensions T and D are determined by the dimensions of the fork ends of the
standard terminals and the area required for bearing of the pin.

If. Dimension K is determined by the required strength and the fork end
dimensions. It shall be neither exceeded nor diminished, except to the extent of the
manufacturing tolerances.

III. Dimension M is a minimum which may be exceeded if desired.

IV. Dimension B is the least width of shank allowable with the given plate
thickness, t, and may be exceeded if desired.

V. When pin plates are used, care must be taken that the stress imposed on the
pin plates by the pin can be carried into the shank of the lug.

(6) In fitting design, care must be taken not to pierce a structural member at a point
that will seriously weaken it, or to make wires or struts unnecessarily eccentric. Proper
allowance shall be made for material removed by bolt holes in designing spars, longerons,
and other members carrying stress. .

(c) Clevis pins should always be inserted in their fittings with the head uppermost
on assembly of the airplane and wire, cable and strut terminals, and bolts should be so
inserted. The fittings should be designed accordingly. This is to reduce the possibility
of the pin or bolt falling out in the event of failure to safety it properly with wire or copper
pins.

(d) The threaded portion of a bolt should not be used to take a shear load when the
shear load on the bolt is greater than 25 per cent of the bolt’s rated shear strength. When
the shear load exceeds this value not over one-fourth thickness of the fitting should bear on
the threads. This condition may be obtained by using washers where necessary.

(e) In determining the sizes of solid steel bolts to be used in wood, the strength of the
wood in bearing against the bolts shall be obtained from Figure 10. No bolt smaller
than the 14-28 size should be used in the primary structure. The use of hollow steel
bolts with comparatively thin walls is to be avoided, as tests at the Forest Products Labora-
tory show that such bolts are little if any more efficient than solid bolts.

Standard eyebolts or similar special bolts having a fillet between the head and shank
with a radius less than 1% inch must not be used in places where they will be subjected to
bending and vibration. Particular care must be taken to avoid the use of such bolts in
locations in the control systems and surfaces where vibration might ‘cause fatigue failure.

(f) Commercial machine screws shall not be used as substitutes for aircraft bolts in
any part of the primary structure.

(g) It is recommended that fittings be made up of the smallest practicable number of
constituent parts so that reliability will be increased and welding, brazing, and riveting
in their manufacture be reduced to a minimum. To attain this end, the use of pin plates26

hen wires of different sizes must be attached to the.

should be avoided whenever possible. WwW ney
d to in order to avoid the necessity for pin)

same plate, spot facing can usually be resorte
plates.
Flame braz

tute parts of the primary structure.
5. Control system.—(a) Where a wheel control is used, the wheel should have an)

angular motion varying from 180° to 360° from the neutral position. The wheel diameter :
should be from 14 to 20 inches. The distance from the wheel in its rearmost position to
the back of the pilot should be 14 inches and its forward travel should be from 18 to 247
inches. The height of the wheel should be such that it will clear the pilot’s legs with the?
seat in highest position, so that the range of seat adjustment will not be limited. The)
rudder pedals on large airplanes may have a motion up to 10 inches total. )

(6) The movement of the control surfaces should be not less than 20° past neutral inj
either direction. In some airplanes it may be desirable to favor an upward elevator)
movement. There should be no interference between the surfaces or their bracing when
one is held in its extreme position and the other operated through its full angular movement.

(c) All control systems should be provided with stops so that their movements will be
limited by these stops rather than be accidental interferences. When possible, such a
stop should be placed in proper relation to the control stick or column near its fulcrum,
so as to limit its movement to the desired maximum in every direction. This will prevent
the control from sticking or bearing against the pilot’s limbs or body in the event of @
structural failure in flight or in crash. It will also serve to protect the instruments and
prevent jamming of the stick behind other controls or instruments.

(d) The maximum pull on the stick that can normally be applied by a pilot in flying
position is assumed to be 200 pounds. The stick should be designed to withstand a ‘push
or pull of 300 pounds in a fore and aft direction and a force of 150 pounds in a lateral
direction. ‘The stick should be made of a nonferrous material, unless the position of the
compass is such that the use of a control stick of ferrous material would not cause deviation
of the needle. A wheel control should be designed to withstand a force of 300 pounds
fore and aft, and a force of 250 pounds applied tangent to the rim of the wheel.

The rudder bar or pedals should be built to withstand a force of 300 pounds applied at
the point of contact with the foot.

(ec) The maximum movement of the control surfaces should correspond to the maxi
mum movement of the controls.

(f) That part of the control system which is between the stops and the control surfaces
should be designed to withstand 125 per cent of the loads specified in Table 1. The parts
between the stops and the pilot should be designed to withstand the loads specified im
paragraph (d). Care should be exercised in the design that friction in controls is reduced t0
aminimum. Properly designed bell cranks, carefully hung pulleys, and on large airplanes,
the use of ball bearings will aid in this.

(g) Unless the design of the control surface hinges is such that the possibility of failure
is very remote, at least three hinges should be used on each surface.

(h) It is important that the linkage be true to prevent binding of some of the parts:
Play or lag in controls should also be reduced to a minimum.

(i) Care must be taken in the design to prevent the possibility of the jamming of the
controls in case of failure of the seat, floor, cowling, or similar parts of the airplane.

(j) When speed is of great importance, cables and control masts should, when practi-
cable, be internal. When this is done, however, inspection holes should be provided,
as it is essential that controls be inspected frequently.

(t) Strength calculations should always be made for the important members in control
systems. In these calculations the strength of all members carrying loads from the control
system to the primary structure should be investigated, and also the possibility of the
arrangements used causing excessive stresses in the primary structure.

ing or partial dip brazing shall not be practiced on assemblies that consti-727

(l) The points of attachment of control wires or push rods to ailerons should be located
opposite drag struts in such a manner as to prevent the application of torsion to the rear
spar. If this is not feasible, the effect of such torsion should be considered in design.

(m) Deductions must be made for bolt holes which pierce a member at a critical
section. If possible, they should be avoided. The stick especially should be investigated.

(n) Where tubes are subjected to both bending and torsion they must be investigated

for the effects of the stresses due to the combined loading. The allowable shearing stress

may be determined from Figure 11.

(0) External bracing of tail surfaces is desirable on all types of airplanes. If unbraced
surfaces are used, the support from the fuselage must be especially rigid.

(vp) The attachment of control masts to surfaces should be given careful attention.
In general, a broad base for the mast must be provided. The masts and ribs to which they
may be attached must be placed in line with the respective control cables. As a rule,
hollow streamline steel masts have been found to give the best results.

(q) In a control system, the attachment of levers to tubes is important. Welding or
brazing alone is not sufficient, but rivets or bolts should be used in addition. The use of a
sleeve to reinforce the tube at the point of attachment of the lever is recommended.

(r) In control systems welds shall in no case be employed to carry tension without
reinforcement from rivets or bolts.

(s) Bushings and pins at bearing points should be large enough to prevent distortion
and consequent looseness and have a means of lubrication to prevent wear. They should
be designed for a maximum bearing pressure of 750 pounds per square inch.

6. Struts and columns.—Long, slender struts shall be designed by the Euler formula,

P/A=c7’E/(L/e)? or P=cr’EI/L’.
Short struts of spruce or steel shall be designed by the Johnson parabolic formula,
P/A=f-fL?/4cr*Ep? or P=fA— (fLA)?/4er° EI.
Short struts of duralumin tubing shall be designed by the straight line formula,
P/A=f—KL/pyc or P=fA—KLA/p-ye.

The quantity c in the above formulas is the restraint coefficient which is to be taken as
1 for pin-ended struts and as 4 for truly fixed-ended struts. In the design of a strut, its
value must be assumed by judgment, but in airplane work, as it is very unlikely that a degree
of restraint greater than that represented by c=2 can ever be counted on, that value is the
maximum which will be permitted by the department. The quantity K in the formula for
duralumin, an empirically derived constant is equal to 400; “Ff” is the yield point strength
of the material except for duralumin, in which case it is 48,000. E,I,L and p have their
usual significance.ir

SECTION III.—_MATERIALS

1. The materials used in aircraft structures must be of the best. Since it is impossible,
at the present time, for the department to draft a complete set of specifications or to inspeet
and approve all materials to be used in aircraft, it is accepting those that conform to the
specifications of the United States Army or Navy, the Society of Automotive Engineers
or other recognized standard. . In any case the use of definite physical properties for a given
material will be assumed by the department as a guaranty by an aircraft manufacturer
that he will use only materials having those properties.

The use of materials of inferior quality or of those which experience has shown to lack
uniformity of quality or strength will be regarded as sufficient cause for withholding approval
of a new design or for canceling approved type certificates or licenses already granted,

2. The following tabulations of strength properties are, for the most part, based on
materials conforming to Army-Navy specifications and they are given solely for the informa-
tion of manufacturers as to what can be expected from materials of a quality satisfactory
for use in aircraft construction. The Department of Commerce does not guarantee them
nor assume responsibility for them in any way but offers them solely as consistent and
reasonable values for materials which may be obtained commercially under standard
specifications.

3. Spruce and ash.—

Properties for airplane spruce and ash conforming to Army-Navy specifications

| Spruce | Ash

 

 

'
SINR es Psy Tcy OTN G EVEL eg ss eer a an ee Se ee ue Gana ee eee ehaataaseun> per cent_- 15 15
AVEO mats Ot OLSSICIGN = sg oe as ee ek ee ee oe a ae akan pounds per square inch..| 1, 300, 000 | 1, 460, 000
Allowable tensile sirdsei:s = isi t ars it PS A a Se ee Bees esd GO: ts 10, 000 | 16, 000
Wodiiiis Of bE eure eClane minal SOCUIONS) = <cc 256855 oo See a Soa ee ce hat heed eSkueae neon do....| 9, 400 14, 800
iEate Com DEeoslod: PAEAI ION COfer Ale es se he So 5 a ee ap ae ean yee sa nos ese do----| 5, 000 7, 000
Witimate compression, perpendicular to grain. 2.2. 02. 23 4 ots go egy i ond oe oe, He. sao -<0 5 O53 840 | 1, 920
IeGUeTEGIMan sna SiTOnU hte Se ge ee Se es Ce Neo eee oe ac ew coe pene dees do... 750 1, 380
PECCInGe ele hGsner GH DIG 1O0bi otc nee ee oe ae ge Se Bett ck Sp mune ae peep esbees pounds..-| oT 41
Average specific gravity (based on oven-dry weight and volume) ___-__./_---------1--------L--------------- see a 0. 40 0. 62
IVETE SDOUIIC Pit VI0y POPUM UCase. Se pote see oe ey Ree SL RA eS ol aos wah esoastccen . 36 . 56

 

Two-ply spruce webs on box spars may be assumed as capable of carrying a longi-
tudinal shear stress of 1,800 pounds per square inch.

 

 

 

 

 

 

4. Steel.—
Allowable stresses in pounds per square inch
& : | Mi | Chrome- c
JSommercial’”’| Mild carbon lmolybdenum Heat-treated alloy steels
Steel No-_--_- SpA Seg ie a i te eee tt | 1, 015-1, 020 1, 025-| 4,130 | 2,330, 3,180, 3,240, 4,130, 6,128, ete,
UilGianate tease strengitian See ee | 42, 000 55, 000 | $0, 000 125, 000 150,000 180, 000
MAO SDOML 2c eres! ial a a et ee ee 27, 000 36, 000 60, 000 105,000 125,000 150,000
| 75,000 90,000 105, 000

Bearing strength 65, 000 95, 000 125, 000 160, 000 190, 000 210, 000

28x108 28x108 | er 29 x 108 29 x 106 29 x 108

HOA DS SITONGL He ae rk ee ce age eee 25, 000 35, 000 60, 000

 

The efficiency of a welded joint in unheat-treated steel should be taken as 80 per cent.
A welded joint made after steel is heat treated reduces the strength properties to those for
the unheat-treated material. Higher values should not be assumed unless the assembly
is heat treated after. welding.

Streamline tube formed from flat sheet and welded is rated as “commercial” and its
strength properties may be otained from that column. Drawn seamless streamline tubing

(28)29

may be obtained in mild carbon, chrome-molybdenum, or nickel steel, and the properties
employed should conform to the type of steel used.

The properties listed for the heat-treated steels will vary somewhat for the different
steels, but the values listed above may be used as a guide.

5, Aluminum alloy (duralumin).—

Weight per cubic foot... -.--.---------------++---+33--+------ +--+ 5-2-5 pounds - - 175
Ultimate tensile strengths. 2522 -2-2-- 222 252.2222 ied ce BE 3 pounds per square inch_- 55, 000
Yield point a tn ti mim ie rm er ae te ee en ee eee ee 30, 000
Poe SOPEMeE APRS OL VCE... ec ik oo eg fe dona teehee GOK cs 30, 000
EEE OEP ORPORE DD. ora ib Sw ee = cide oo open db Ene ab Ses ae ie de he Bee a 75, 000
Modulus of Mes - be le ee ae bad KG hae § Seka heck aoe e do.--- 10, 000, 000

The properties listed above will vary with the different alloys, but those given may
be used as a guide for the more commonly available alloys, such as 17ST, made by the
Aluminum Co. of America.

 

FreelLooding Of upper Wwihg

Loading on Jower

 

horio

WING

30

    

Chord of upper Wing

 
   

ww
Ss

Luion of heading Biplone wing
Kecommended Slancard
for of! angles of attack.

    
 

90

  
   

        

 

70 35 -30° 10° 0? "0° 720 *30°
stagger
Ria 2
vag EGE
cen ng x“
Wing Tip D i Se
A G NOTE: For _externaly braced wings

 

DE is at ourer Struts

for internally braced wings

- QE 1s solecated thot OF = BF

—Ly “Tip Length——|F oe
/-Cot<-=CoFA

Sem Span =

wr

 

 

 

 

 

ty
s

Leading & oe
FG EM

 

Plane of Symmetry or side of Fustlege>

Fie. 331

 

 

 

 

 

 

Wing Tip

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Se= Cffective Semi- span

 

Ly = Tip Length

 

5 = Sermi- span i

 

v
fers ay
4 % RS 3 <
f 8 ‘SI'S ys Ys .
a 5 NES VS .
“Sl s S| gS ¥
we OF Sia Ix S
a g! RIS |S 8} § ng
| Sw IS i 8
Of Oy “IS ak >
3 R
;
wd/= Lead weight of wings g
L
rote Se ~ Effective Sem!-spon ‘a
p—— “4 = 770 Lenglh—er .
5 = Sem) - spon ——___l8
Fia. 4
ee r 8 2 i
A « & 8
fs g oe & x S| $
Vik ys HE, ae a 8
Q yy
oo. 2 SH NI Ss -
wo = $}S y
WH Ir SLs zg & Ne
S - SIS ab & g <
SHE $ |
Ny wed = Dead weigh? of wings g
4
Se
3
§
S

 

4 fara

Fiq. 5

 

ig

Projecting portion of aileron
“ GSB he

 

 

 

 

 

 

YY
VY Aiileron
H
Z ‘as
~ ;
Note-AG=GB & : Wing proper
NR
Six
aS

 

 

 

Tip of Wing proper

A-ALAN OF WING ;
Showing projecting portion of aileron by cross halching

 

 

le scorns
fa £ NOTE: |
o GE & EF are straight lines

x F /s at the ovler pont of
¥ Inflection of the outer bay
J
i

G a

B — Curve of added benaing noinen? or +€ar Spor
due to ladon projecting Porv/or? Of alr Or

Fia, 6a2

 

i
= MI OS Id
connod 40 SoMeSHOHLN Y SSFLIC LIND ITIL TAP

oa \9

Fia. 7

WIDTH OF BEAM

THICKNESS OF WEB b
5PRUCE BEAMS

VRATIO

4

oe33

DISTANCE BETWEElY POINTS OF /NFLECTION
ON SPAR WITH RIAL COMPRESSIVE AND
AND UNIFORM 1 ee LOADS.
1Y1-2 z eae Ef
7? = Stee ha 7 P
Me Me

6 |p ¥ E130

mye
Ss

28

it
TMT
~
8

Ss
Lili
g

 

tuul the

Dp MANS
< S Q
2s Ss
&

MMT TM
P

>

 

 

 

3002 ; es ERE ol 120
2505 Mr1-2

S

TUTTI TITTY TT

mS
a
8

(r
200
190
180
170
160
(50
190
130
120

~
3

\o

S

N“
8

 

tli eta beta batted a

un TOT TE
8

co oO NN
SS - Ses
Oe Se eS

 

4 60
= 0 °
700 700 i
a 90 50
Mz-2 600 Pe
Ww 7 a oad &0 a
450 = 60 -
.400— = .
a — 50 2
350 = =)
mo — FO e
-300 = E tee
a 30 E
200 oF e
~ E 20
.200— oe bs
/50— c

 

 

s
a

/0
Fia. 8

seis34

4500

COMENTRIC
LOADING

4000

3500

3000

2500
”
ag
§§
Q
2000
BOLT
BEARING STRENGTH OF BOLTS
IN SPRUCE
Co NTKC LOADING AT

STRFES.
ae PULL ON BOLT.
‘ AINGLE WITH THE GRAIN OF THE
W000 LESS THAN22° USE THE VAL
1000 FROM THE CHORT 8 THE
OF THEPULL ON THE BOLT PARALLEL
THE GRAINY. WHEN THE ANGLE /s
GREATER THON 22 VSEO9 THE VAL
FROM THE CHART &THE
500 OFTHE Pull PERPENDICULAR To THE
: GRAIN. FOR ECCENTRICLOADING LI
RESULTS FROM THESE CURVES BY _Z.

£0 OM FPL, L-228-7 FOR ASH WWLTIPLY BYLG FOR
&FRIER F217 . BY.7 FOR

 

 

  
   

o

W/DTH OF WOOD MEMBER !N /IKHES
Fig. 10No. Roted 2. i ™M e 7 : : ‘ Width of Shank 8
streng = h2| ‘leo | | %2 | %lwlAln |%
| 4100 |7h6|"%al bz |Ke|%2| 41 21H! %
21 |2,100 |" '42|%2|hol/e |! 4o| | 7a) 2) %
34 [3,400 | 7%. |/342| a2] 6 | Welz Mo |/ Meo |/ Me| "he | Bl %
46 | 4600 | he|'%452| We |\%z He 1G \1Yoli Ho | Hel Ke
6/ | 6/00 | 7a \*%al Ae | %el "Neo 2% |\2 ol1% \1 te | Va
80 | 8,000| 7a |*%al %e |\%al A 2k Blais | %
125 | 12,500 |"hol’He |* 42 |G |"Mo 2 4e\2  \1 %\1 He| %
175 17,500 | fa |'%e | 7/4 Phe |% 2% \1 4 |! Goll he | %
Fig. 9
70 VEE CAR ae Fa a ee a ee
= SI
6ok—y _
pe} :
Ss0F SN _
a 4
ec. £ NN SPEC|2330 a
SF “
i 4
3” - rN Se
=f NM a
ae a :
= = roe it
Lo L SPEC | (025 2
¥ 0b S q
ee I Poca q
Sf Sj
<r DURALUIMIN fr actin a
S25 =
20 PS zt
a. t od =
- — a
S z
S & es
A a
ee
Sor :
~~ ene —_
QF eos
2 ~
obithiitiiis $4324 £44 oy EECLer EE He z
RATIO ¥
Fia. 11

O