Aeromodelling

AEROMODELLING

■
R. H. WARRING
■% •

Author of Radio-Controlled Models, etc.

LONDON
MUSEUM PRESS LIMITED
First published in Great Britain by Museum Press Limited
26 Old Brompton Road, London S. W.j
1965

© R. H. Warring, 1965
CONTENTS
t

CHAP. #

1 Model Aircraft Defined and Classified . . 9

2 Tools and Materials . . . . . 17
..

3 All-Balsa Models ..... 29

4 Tow-Line Gliders ...... 37
5 Rubber-Powered Models .... 46
6 Power Models ...... 59
7 Jet Models ....... 68
8 Unorthodox Models . . . . 76
9 Construction ...... 87
10 Covering and Finishing . . . . .102
11 Trimming and Flying . . . . .111
12 Engines . . . . . . . .122
13 Control-Line Models . . . . .134
14 Radio-Controlled Models . . . .144
15 Indoor Models . . . . . .154
16 Contest Models . . . . . .160
Index . A
. . . . . . . . 167

PRINTED IN GREAT BRITAIN

BY EBENEZER BAYLIS AND SON, LIMITED
THE TRINITY PRESS, WORCESTER, AND LONDON

R-3341
 TABLES
l

1 PAGE

I. Aeromodelling Tools 18

II. Balsa Grades ..... 20

III. Typical Weights of Sheet Balsa 21
IV. Typical Balsa Selection 21
V. Plywood Thicknesses .... 24
VI. Design Proportions for Rubber-Powered
Models ...... 51
VII. Typical Rubber Motor Sizes 53
VIII. Maximum Safe Turns for Rubber Motors 56
IX. Typical Sizes of “Duration” Power Models 65
X. Typical Sizes of Semi-Scale Sports Power
Models ....... 65
XI. Typical “Jetex” Model Sizes 7i
XII. Typical Ducted Fan Model Data 74
XIII. Typical Fan Sizes . . . . . 74
XIV. Design Data for “Jetex” Helicopters 79
XV. Summary of Material Sizes, etc., for all Types
of Models ..... 97-101
XVI. Guide to Application of Covering Materials 109
XVII. Typical Rigging Details for Free-Flight Models ”4

XVIII. Trimming Chart for Free-Flight Models 120

120
7

i
8 TABLES

XIX. Standard Engine Sizes .... 124

XX. Equivalent Capacities .... 124
XXL Approximate “Equivalent” Engine Sizes 126
CHAPTER I
XXII. Typical Sizes of Sports Models and Trainers . 142
XXIII. Typical Propeller Sizes for Stunt Models. 142 MODEL AIRCRAFT DEFINED AND CLASSIFIED
XXIV. Typical Sizes of Speed Models 142
The general description “model aircraft” embraces a whole
XXV. Radio-Controlled Model Characteristics variety of different types, with a very wide range of interest and
152
XXVI. Team Racer Specifications .... appeal both as regards age and intelligence level. Thus many
163
of the more elementary types of model aircraft, some of which
XXVII. Control-Line Speed Classes .... 165 may only cost a shilling or so, are properly classified as “toys”
and are specifically intended to appeal as such. At the other end
of the scale there are model aircraft built with maximum per¬
formance in mind, and conforming to particular specifications,
as a world-wide hobby-sport; and radio-controlled models
which may cost up to £30 to £35 to build and in addition carry
electronic equipment itself costing a further £100 or more.
Their appeal is essentially adult, although many classes of
contest-type models are basically inexpensive, placing a pre¬
mium on individual skill, and appealing to all types of en¬
thusiasts from the middle teenage upwards.
The first main broad sub-division of model aircraft types is
into flying and non-flying models. The latter are invariably scale
models, ranging from simple assemblies of carved or moulded
components (e.g. plastic kits) to fully detailed and highly
elaborate models following similar “skeleton” or framework
details as the full-size prototype. Primarily they satisfy a
collector's instinct and cover a quite distinct branch of aero-
modelling.
The purpose of this book is to cover flying models, and here
again we can immediately split this main group into free-flight
models and control-line models. The former covers all models
capable of flying freely through the air in a similar manner to a
full-size aircraft. Control-line models, on the other hand, are
restrained in their flight path by being attached to lines (usually
two but sometimes a single line), and so must fly in circles.
Actually their flight path is not as limited as might appear at
9
 IO AEROMODELLING MODEL AIRCRAFT DEFINED AND CLASSIFIED II

first sight since one of the basic features of a control-line model (iv) Jets—where the power unit is a miniature jet engine (in
is that the operator actually controls elevator movement during practice, confined to standard sizes of a small proprietary
flight, permitting a wide range of manoeuvres to be carried out. rocket unit called “Jetex”).
Free-flight models have the greater appeal as flying models, Each of these groups may embrace three distinct types, re¬
but require fairly large areas of open ground from which they ferring basically to overall shape and appearance. These are,
can be flown. Control-line models can be operated in relatively original designs where shapes and proportions are derived solely
small areas and also place far less premium on design and trim¬ on the basis of achieving maximum performance; scale models;
ming skill, since they are flown under control. They are also and semi-scale models. Original models, especially those designed
less affected by winds for the same reason, and generally are

Fig. i. Semi-scale glider layout features a clean and generally “realistic”

appearance, although in this case a polyhedral wing is employed.

more robustly constructed in any case since weight is not such

a. critical factor. The two basic forms—free-flight and control-
^ne thus have quite different appeals. It is unusual to find an Fig. 2. A low wing semi-scale rubber-powered model. Although “realistic”
in appearance it is a tricky model to design and fly.
established modeller enthusiastic about both. He usually con¬
centrates on one type of flying or the other—and generally
for contest flying, generally look quite unlike full-size aero¬
specializes in a particular type. There are so many different
planes, and are often criticized in this respect. There are many
types to choose from that no one person can hope to become
good reasons why they should have such unusual shapes or
expert with each, although a lot of useful experience and
layouts, however, all connected with stability and efficiency.
knowledge can be gained in trying different types initially to
A full-size aeroplane is flown under the control of the pilot,
find the one which offers the most satisfaction or interest.
who can correct any deviation from a proper flight path caused
Free-flight models can be grouped according to their motive
by gusts, etc. A similar form of “control” has to be achieved
power, as—
automatically with a free-flight model—and since a model is
(i) Gliders—which have no motive power (but many gliders
so much smaller and lighter it is more readily upset in any case.
are convertible to auxiliary power by the fitting of a small
Thus a successful free-flight model has to have a large reserve
pylon-mounted engine).
of built-in or inherent stability, which can be achieved by
(ii) Rubber models—where the power to drive the propeller
such methods as increasing the upward inclination of the wing
is supplied by the unwinding of a rubber motor (never “elastic”
tips or dihedral (and in some cases producing the required
that is purely for toys, and many rubber models are far
dihedral in two stages by cranking the wing); increasing tail-
from being toys!).
plane and fin areas; using a fairly long fuselage to increase the
(iii) Power models—where the power unit is a miniature moment arm between wing and tail; mounting the wing above
internal combustion engine.
the fuselage on a pylon; and so on.
 12 AEROMODELLING MODEL AIRCRAFT DEFINED AND CLASSIFIED 13

From the efficiency point of view, wing sections are usually the designer aims, basically, to produce an original layout with
made thinner and undercambered; fuselages are reduced in good stability and performance and then add “realism” to the
section to save drag and weight; and, in the case of rubber shape, such as by incorporating a cabin in the fuselage shape
models in particular, a larger diameter propeller is used. Thus and generally aiming for attractive or “realistic” wing shapes
between the demands of stability and efficiency the resulting
design bears little or no resemblance to a full-size aeroplane

Fig. 4. Typical of the older pylon layout which originated about 1940 for
power duration models. Modern pylon models are even more functional
with thinner, stick-type fuselages.
Fig. 3. This is a functional rubber-powered model which still maintains
something of the general appearance of a full-size aeroplane, A “stilty” and proportions, etc. The greater the “realism” introduced, in
undercarriage is inescapable with a rubber-powered model.
\ general, the greater the penalty as regards both stability and
performance. No such model can hope to have a contest-type
and the more specialized a contest design it is the greater may
be this difference. It is, in fact, no longer a model aeroplane but performance, yet at the same time it will be safer to fly, and
a small-size aeroplane designed to fly without a pilot and to fly much better, than a scale model.
As far as free-flight models are concerned, therefore, it is
give the best possible performance within the limitations set by
its size and type. In other words, it is really a “full-size” impossible to have it both ways. The high-performance original
pilotless aeroplane!
Conversely, since the original design needs so much depar¬
ture from an orthodox full-size aeroplane layout, it is to be
expected that the scale model which duplicates a full-size outline
■)
will not make a good free-flight model. This, as a generalization
is true. However, certain full-size layouts have a reasonable
margin of stability to start with (notably high-wing mono¬ Fig. 5. This layout is a compromise between “performance” and “looks” in
a free-flight power model—retaining a pylon-type wing mount but turning
planes), and can make suitable flying scale models with some
it into a cabin.
further modifications—generally an increase in wing dihedral
and tail surface areas. Although they may be made reasonably design cannot be made to look “realistic”; whilst the wider
stable in flight by this (or other) means they will still not have appeal of the realistic scale model is offset by its poorer flying
anything like the performance of an original design. Even to performance and its greater liability to crash and damage itself.
approach a similar standard of performance certain design Basically, therefore, it is necessary to decide which type of
features must be so exaggerated that the model no longer flying you want. If you want good flying performance, then an
approximates to anything like near scale. original design is the logical choice; although a semi-scale
This leads to the third basic type—the semi-scale model. Here model may be more attractive if you are only flying for fun
 14 AEROMODE LLING model aircraft defined and classified 15

rather than aiming for “contest” performance. Should a scale to provide even more rapid response to control movement and
model be the choice, then it can only be emphasized that are intended to be flown two or more in the same circuit by
previous experience with original or semi-scale models will be independent pilots in “combat.”
invaluable in tackling the trickier trimming problems involved. (iv) Team Racers—which are essentially semi-scale models
The majority of flying model enthusiasts make the basic mis¬ on the lines of full-size lightplane racers for racing two or more
take of starting out with a scale design as their first model, in the same circuit over a specified distance (number of laps).
because the realism appeals to them, when they are likely to (v) Speed models—which are designed for out-and-out speed
achieve only disappointment and an early end to the model. performance and are classified by engine size.
This applies particularly to the smaller sizes of models. Larger

Fig. 6. The unusual layout is popular for power sports models, but unsuited Fig. 7. The typical “sports type” free-flight power model has a cabin and
for high-performance contest flying. “realistic” outlines and proportions.

scale models of a suitable prototype are generally easier to Scale models are sometimes classified separately, although
trim, particularly in the case of power models. they are essentially “sports” type models and seldom have a
Control-line models are invariably engine-powered. No fully aerobatic performance.
rubber motor or “Jetex” unit would give a long enough flight, Both plans and kits of all types of free-flight and control-line
although pulse-jets may be used on control-line models (see models are readily available. Until a certain amount of practi¬
Chapter 12). Stability requirements are reduced to a minimum^ cal experience has been obtained with a particular type of
so in this case scale models are just as suitable as original model it is generally best to build from a kit, particularly as the
designs for normal flying. However, to extend the scope of modern kit usually contains a proportion of prefabricated parts
control-line flying—e.g. to make the model fully aerobatic (e.g. die-cut sheet parts) which make for greater accuracy in
(within the limits of the model being tethered), or for maximum assembly. Instructions are usually specific both as regards
speed performance, again original design features are intro¬ assembly and trimming for flight. The more experienced
duced. These considerations lead to the following types of builder may prefer to work from plans and basic materials,
control-line models— choosing published designs of particular appeal or proven per¬
(i) Sports models—which may be scale, semi-scale, or original formance. Not until the modeller is experienced both in con¬
in layout and have limited aerobatic performance. Original struction of and flying a particular type should he attempt to
designs with a reasonable aerobatic performance are also called design his own models.
“trainers.” The “traditional” method of achieving the best possible
(ii) Stunt models—which are essentially original designs results with free-flight models was to start with gliders, progress
evolved to provide maximum manoeuvrability. to rubber models and thence to power models. This no longer
(iii) Combat models—which are really stunt models designed holds true as with good kits available, satisfactory results may
 i6 AEROMODELLING

be obtained by starting with a semi-scale power model. Trim-

ming and flying technique, however, is something which can
only be mastered with practice, and there are no short cuts in
this respect other than choosing a good basic design to start
with. This is a model with a good reserve of inherent stability chapter 2

and a tolerance to being mis-handled, rather than aiming

TOOLS AND MATERIALS

A particular virtue of model aircraft construction is that it

requires only a minimum of tools and equipment. The one
essential feature for a “workshop” is a flat and true building
board at least as large as the longest single component of the
model under construction. A large drawing board is excellent
for this, but unnecessarily expensive if purchased new. Any
similar board will do—or a softwood kitchen table top pro¬
Fig. 8. For free-flight scale models a high-wing design makes the best
prototype. vided it is flat and true, and is soft enough for pins to be stuck
into it. The building board must be true and smooth-surfaced,
straight away for a model which has maximum performance, for flat frames are built directly on it and their accuracy
or high degree of scale realism. depends on the board surface being true. It should be kept only
With control-line models, success is even more a matter of for model building.
practical flying experience, starting with a “trainer” to master Also required is a working surface on which the board can
the rudiments of controlling the model in circling flight and be rested. This should be at least three feet long so that standard
then advancing to a more specialized design. lengths of sheet balsa can be cut on it. Alternatively, the build¬
Time spent in building and flying a “trainer” type of model ing board itself can be used as a base for stripping and cutting
in any category is never wasted, when starting aeromodelling, wood where the working surface is a piece of domestic furniture
as success is only built on experience. Basically, in fact, the which must not be damaged. It is a little difficult to use a
hobby of both building and flying model aircraft is essentially building board for both purposes, however, once a plan has
a practical one, and a real knowledge and appreciation of been laid on it and building has commenced.
the problems involved starts only when the first model is A basic set of tools boils down to a modelling knife with
completed. spare blades, some razor blades, a razor saw, a steel rule, a
hand drill, square- and round-nose pliers, and a fretsaw. Other
tools will be found useful, but not necessarily essential (Table I).
The modelling knife is used for most of the cutting jobs in
sheet and strip balsa, and can also be used for carving with a
different blade. Typical blade shapes and their purposes are
shown in Fig. 9. These represent only a proportion of the blade
shapes available, but the others are more suited to wood
carving jobs rather than model aircraft construction.
The modelling knife can be used for cutting strip balsa to
l8 AEROMODELLING TOOLS AND MATERIALS 19

length in all sizes up to about 3/16 in. square. For larger sizes,
TABLE I. AEROMODELLING TOOLS and particularly in denser wood, the knife blade tends to crush
the wood and it is difficult to produce an accurate cut, so a
Tool Use(s)

Modelling knife. Cutting balsa sheet and small strip sections,

carving and shaping.
Razor blades. Cutting thin balsa sheet and strip, trimming
tissue covering.
Razor saw. Cutting larger strip sizes.
Stiffback saw. Cutting balsa block.
Fretsaw. Cutting thicker balsa sheet to curved outlines,
cutting ply.
Coping saw. Cutting thick sheet balsa and block to curved
outlines. Fig. 9
Hand drill. All drilling jobs; also, fitted with a hook, for
winding up rubber motors. razor saw is a better tool (Fig. 10). This can also be used for
Square-nose pliers. Making “square” bends in wire, holding nuts
accurate straight cutting of thicker sheet balsa and block.
when tightening, etc.
Round-nose pliers. Bending wire loops, etc. A razor blade can also be used instead of a modelling knife
Vice. Bending heavier gauge wire, holding parts for for sheet and strip balsa. It is not so accurate for cutting thicker
drilling, etc. sheet, but is generally more accurate than a knife for cutting
Soldering iron. Soldering wire parts together. (Note: for 1/32 and 1/16 in. thick sheet. Razor blades, in fact, used to be
radio-control model installations, an electric
soldering iron is essential.)
Small clamps. Useful for certain assembly jobs, but most
clamping requirements can be met with pins.
Small screw¬ For nut and bolt assemblies (e.g. mounting
drivers. engines).
Steel rule. Measuring and also as guide for cutting
straight lines with a modelling knife.
Sanding block. For all sandpapering jobs, although a piece of
thick balsa sheet will usually be suitable instead
of a special sanding block.
Hacksaw (small). Metal cutting; also useful for cutting off
engine mounting bolts, etc., and thicker
gauge wire. the main cutting tool for aeromodelling work before modelling
Wire cutters. For cutting steel wire up to 16 s.w.g. knives were introduced. Today, however, they are mainly
Files. For cutting engine mounting bolts or steel
reserved for lighter cutting work, and for trimming off surplus
wire thicker than 16 s.w.g.; also for balancing
plastic propellers. material after tissue covering.
Hand spray. Invaluable for water-spraying tissue covering. For cutting such plywood parts as may be used a fretsaw is
Spraygun. The best method of applying dopes and necessary, although a razor saw or small hacksaw can be used
finishes, but brushes can be used instead. for straight cuts. A fretsaw or coping saw is also the best tool
for curved cuts in thick sheet balsa (e.g. \ in. thick or more).

\
20 AEROMODELLING TOOLS AND MATERIALS 21

Pliers are used for wire bending and cutting (although wire TABLE III. TYPICAL WEIGHTS OF SHEET BALSA
thicker than 16 s.w.g. is best cut with a small triangular file). (weight in ounces for 36 X 3 in. sheet)
Small flat-nose pliers are also extremely useful for inserting and
withdrawing pins used to hold parts in place during cementing Thickness h in. is in- sz in. £ in. W in- i in- fin. £ in.
up. Pins themselves are an indispensable part of model building.
Some people prefer glass-headed modelling pins with hardened less less less less less less less less
shanks, others use ordinary domestic pins. Whichever type you than than than than than than than than
choose, buy by the gross in order to be sure that you always Ultra-light 1 & 1 l£ 2j 3
have enough. Light f is l i£ 4 2i 3
Medium (typical) t i i£ 3 \\ 6
TABLE II. BALSA GRADES
Hard £ I 4 2 3 4 6 8
over over over over over over over over
Extra-hard £ 1 i£ 2 3 4 6 8
Ultra- Medium- Hard Extra-
Grade Light Light Soft Medium (or Hard
Heavy)
TABLE IV. TYPICAL BALSA SELECTION
Density
lb./cu.ft. under 6 Component Grade Cut
6-7 7-9 9—12 12-16 over 16

Fuselage: Longerons Medium or Hard Straight grain

Spacers Medium or Medium-Soft Random
The standard material for model aircraft construction is
Sheeting Medium-Soft Random
Balsa. This is a tropical wood found only in South America and Random
Wings: Leading edge Light
which, due largely to its rapid rate of growth, is very much Spars Hard Straight
lighter than any other wood. At the same time it varies widely Trailing edge Medium Quarter grain
in density from tree to tree, or even in the same log. It is usually Ribs Light Quarter grain
graded according to density or weight per cubic foot (Table II), Sheeting Light Straight grain
strength being more or less proportional to density. Thus hard Tailplane: Spars Medium-Soft Quarter grain
(heavier) balsa is used for parts which need the greatest Ribs Light Quarter grain
strength (such as wing spars) and softer (lighter) wood for Solid Block Tips Ultra-Light Random
parts where weight is to be saved. Plans often specify which Sheet Balsa:
Wings Light Quarter grain
grade of balsa is to be used for particular parts, otherwise
Tailplanes Light Quarter grain
modellers choose the grade according to their individual
Fins Light Quarter grain
preferences. Failing such experience, Table IV can be used as
a guide.
The actual properties of a length of balsa (particularly balsa and tend to split if bent. This is normally termed “quarter-
sheet) will also depend on the “cut.” If the sheet is cut from a grain” stock and can be identified by the speckled appearance
log tangentially to the annular rings it will be fairly “bendable” of the surface grain. Quarter-grain balsa is the logical choice
in an edge to edge direction. Sheet cut from the log in line with for parts which have to be stiff and rigid, such as wing ribs and
the medullary rays, on the other hand, will be much stiffer sheet tailplane surfaces. Tangent-cut sheet is the logical choice
 TOOLS AND MATERIALS 23
22 AEROMODELLING

for curved components, such as sheet covering for wing leading thin and a small section spar has to be used. It is roughly three
edges. times as heavy as medium-grade balsa, but very much stronger.
The basic “cuts” are illustrated in Fig. n, whilst Table IV It may also be used for other highly stressed members in other
gives their typical applications. The experienced modeller often built-up structures on larger models, e.g. longerons. The hard¬
goes to a lot of trouble to sort out the right “cut” as well as woods, such as beech, ash, etc., are used exclusively for motor
“grade” of balsa for a particular model. The less experienced bearers on power models, and for other “strong” points such as
modeller can receive guidance from the local model shop where undercarriage mounting points on radio-control models where
he buys his stock materials. In the case of kits, “grade” and the main undercarriage may be attached to the wings.

TANGENT CUT

RANDOM CUT

'QUARTER,

Plywood is used for main fuselage formers, such as the front

“cut” selection is done by the kit manufacturer, but is by no
former or firewall on power models which carries both the
means as complete as can be done by individual selection. In
motor bearers and undercarriage fixing. It is also used for local
some cases, too, choice of grade is influenced more by pre¬
reinforcement, such as spar braces at wing dihedral joints, etc.
fabrication requirements than anything else. The good kit
Thin ply, fretted out, may also be used for wing ribs on
design, however, allows for this and other possible variations.
Continental model designs where balsa may be comparatively
The experienced modeller building from a kit may, however,
scarce, but ply is not normally regarded as a substitute for
prefer to replace certain parts—particularly wing spars, wing
balsa sheet, only as a material with specific applications.
tip blocks which may not be equal weight, wing sheeting which
The best plywood for model aircraft use is aircraft quality
is too heavy or too rigid, etc.
beech ply, which is available in thicknesses from 0-5 mm.
Other woods used for model aircraft construction are obeche,
upwards. Ordinary “domestic” plywood, or marine ply (which
spruce, hardwoods such as ash and beech, and plywood.
is a mahogany ply), is not generally suitable or available in the
Obeche has relatively little application, being as heavy as
thinner thicknesses. All plywood thicknesses are specified in
spruce and harder, but more brittle. It was used as a substitute
millimetres (except in the U.S.A. and Canada) although
for balsa when that material was in short supply during World
commonly spoken of in terms of nominal inch thickness in this
War II and has survived as a generally modelling wood (par¬
country (Table V).
ticularly for boats). Today it has little or no application for
Limited use is also made of other wood-like materials, such
model aircraft construction. Spruce has a particular applica¬
as bamboo and reed cane (Fig. 12). Both are materials which
tion for spars in larger model gliders where wing sections are
 24 AEROMODELLING TOOLS AND MATERIALS 25

can be readily bent to curved shapes for wing-tip outlines, etc. model construction is a comparatively new material—ex¬
by gentle heating in the case of bamboo, or simply bending in panded polystyrene. This, basically, is a “foamed” plastic which
the case of the flexible reed canes. Bamboo, being a rigid, can be expanded in a mould to a required shape, or into slab
strong material, is also used for rear pegs on rubber motors form for subsequent cutting to shape. Material density may
(anchoring the rear end of the rubber motor) and is much range from as low as 3 pounds per cubic foot up to 10 pounds
better than a hardwood dowel in this respect; and for rubber per cubic foot, a typical density for general purpose mouldings
model undercarriage legs. being about 4 pounds per cubic foot. It is thus lighter than
balsa but, unless produced in the form of shell mouldings, must
TABLE V. PLYWOOD THICKNESSES be shaped “from the solid.” Thus an expanded polystyrene
component such as a wing is not necessarily lighter than a
built-up balsa wing.
Nominal Approx.
Thickness Equivalent
Kit models based on expanded polystyrene construction in¬
variably supply the parts involved as finished mouldings. The
0-5 mm. _ strength of such mouldings may be enhanced by a process of
0-8 mm. is in- remelting and pressing the surface to form a toughened skin
1 mm. & in. (usually a part of the moulding process). Home-made expanded
1 ^'mm. w in- polystyrene parts can be cut from solid with a heated wire, the
2 mm. A in. material being too soft and crumbly to saw or carve in the
2 •g'mm. is in- usual way. “Skin-toughening” can then be achieved by cover¬
3 fmm- i in. ing with tissue or, in the case of larger components such as
4 mm. is in- wings, covering with sheet balsa. The latter technique provides
5 mm. If in.
mm. a rapid method of making large wings for radio-control models.
6 J in.
Both the making and finishing of expanded polystyrene parts
require special techniques, such as special adhesives and fillers,
and also protection of the surface against model engine fuels
Sheet plastic materials have very limited application, except when used for power model construction. Normal cellulose
for “transparencies” and mouldings. The material normally dopes and finishes cannot be used on polystyrene as these will
used is acetate sheet for “glazing” cabin windows on scale or dissolve the material.
semi-scale models; or moulded to form cockpit canopies, etc. Glass fibre mouldings are also used to a limited extent for
Mouldings in thicker acetate sheet may also be used for cowl- larger models. The main disadvantage is that glass fibre
ings, etc. Thin “Perspex” sheet may also be used for larger mouldings are relatively heavy, compared with balsa construc¬
cockpit canopy mouldings, etc., but acetate is the normal choice. tion for example, and thus only offer advantages in larger size
Kits normally include finished mouldings of this type (which thin shell mouldings where the high strength/weight ratio of
merely need trimming to fit); and may also include cowling the material may be employed to full advantage. Such appli¬
mouldings and similar detail parts. Ready-to-fly models are cations include the moulding of fuselage and even wings for
also produced from polystyrene sheet mouldings, this material larger radio-controlled models; and control-line models of the
being more dimensionally stable than acetate. It is not, how¬ speed or Team Racer type where weight is not a critical
ever, normally used for home-made mouldings. factor. It may also be used for smaller mouldings, such as
The only other plastic material used to any extent in flying power model cowlings, etc. Considerable use is also made of
 7
26 AEROMODELLING TOOLS AND MATERIALS 27
glass fibre as a reinforcement material applied over sheet balsa again is a cellulose product rather like a very thin balsa
construction, such as in the strengthening of the front of a cement), or dextrin-type pastes like photographic mounting
fuselage on a power model. paste, “Bondfix,” etc. Choice is usually a matter of individual
The standard adhesive for model aircraft construction is ,1 preference, the pastes being easier to use since they do not dry
balsa cement. This, actually, is a cellulose adhesive which has as rapidly as cements or dope, although the latter may make a
the particular virtue of setting rapidly to produce very strong neater job.
joints with porous woods. Its characteristics can be controlled Only one other type of adhesive need be mentioned—the
by the amount and type of solvent, addition of resins to two-part epoxy resin capable of sticking virtually anything to
strengthen the cement, and so on. Thus there are various types anything. This is expensive, but enables glued joints to be
of balsa cement, some very strong (and usually slower drying), made between materials which would otherwise have to be
others less strong, but perhaps with faster drying and setting. soldered, welded or bolted or riveted in place—e.g. for the
A “strong” balsa cement is suitable for gluing all woods. Other » attachment of small metal fittings, etc. The best-known adhe¬
balsa cements may be excellent for use with balsa but less sive of this type is “Araldite,” which is a particularly useful
satisfactory on other materials, such as ply. Thus a strong balsa addition to the model workshop for detail assembly jobs, etc.
cement is a good general purpose choice. Some strong cements, Covering materials for built-up structures include tissue
however, contract considerably when drying and may warp papers, silk and nylon. The tissues are special papers, not
or distort fragile balsa structures. In such cases it may be ordinary tissue. The lightest is Japanese tissue, particularly
necessary to use another type of balsa cement. suited for the smaller, lighter models. Model tissues, somewhat
Another type of adhesive which has come to the fore for coarser in texture but stronger, have been developed in a
model aircraft work is PVA, popularly described as “white number of grades, but mainly “lightweight” and “heavy¬
adhesive” or “white glue.” This gives excellent strength with weight.” The former are comparable in weight with Japanese
all woods, does not contract or warp structures on drying, and tissue but absorb more dope in finishing, so produce a heavier
does not form hard blobs or smears of surplus adhesive like overall covering. Their scope is virtually the same as Japanese
balsa cement. Smears disappear on drying, and surplus cement tissue. Heavyweight model tissue is used for covering larger
can easily be wiped off after completing an assembly since models.
PVA takes considerably longer than balsa cement to dry. This Tissues represent the lightest covering materials for outdoor
longer drying time can be an advantage when gluing up large flying models, but are relatively brittle. Thus they are easily
surface areas, such as sheet covering, but increases construction split or torn. Stronger materials, such as lightweight silk and
time on other assemblies. PVA is an alternative to balsa cement nylon, are often preferred for larger models. These invariably
for almost all model aircraft gluing jobs, but is employed weigh more than tissue coverings when doped and are also
mainly on larger models. unsuitable for application over lighter structures since tauten¬
Special cements are used for gluing plastics—e.g. poly¬ ing under the action of the dope will produce warps. Nylon
styrene can only be glued with polystyrene cement (usually is considerably superior in strength to silk (which like tissue
called “plastic cement”); a thinned polystyrene cement or can become brittle with age and is not all that resistant to
PVA is used for gluing expanded polystyrene; Perspex cement being split). Nylon chiffon is a favoured material for covering
is used for gluing Perspex. Acetate sheet, being a cellulose radio-control models and others of about 48 in. wingspan
plastic, can be glued with balsa cement. upwards. In selecting nylon for use as covering, however, it
For attaching tissue, silk or nylon covering, tissue cement should be remembered that the lightest material is not always
may be used (a thinned-down balsa cement), dope (which the best since the more open weave may demand a considerable
28 AEROMODELLING

number of coats of dope to fill, each coat increasing the weight

of the covering.
Other materials used for the finishing of model aircraft
are described in Chapter io.
chapter 3

ALL-BALSA MODELS

The most elementary form of model aircraft construction is

that employing solid construction—that is, wings, fuselage,
and tail parts are cut from solid balsa sheet, then shaped as
necessary before being cemented together to form a complete
model. There are definite limits to the size of such models,
beyond which the weight of solid construction becomes pro¬
hibitive. However, “solid” construction may still be retained
for components on larger types of models, such as wings and
tail surfaces for control-line models, and fins for power models.
The chief attraction of such construction is simplicity, with
building time reduced to a minimum. Also the resulting
structure is quite rugged.
The main classes covered by solid balsa construction are the
toy flying models and chuck gliders. Toy flying models may
be gliders (essentially similar in form to a chuck glider, and
often so-called) (Fig. 13); or rubber-powered models (Fig. 14).
In the former case size is usually limited to a span of about
3« AEROMODELLING ALL-BALSA MODELS

may either be moulded in, or produced by cementing the

wing panels to curved ribs.
The true chuck-glider, although still largely a “toy,” is
larger, with wings cut from thicker sheet balsa and carefully
shaped to an efficient aerofoil section—Fig. 15. The model
size is also larger and extreme care may be devoted to providing
a fine, smooth finish by filling the grain of the balsa, rubbing
down, and polishing—all aimed at improving efficiency.

Fig. 14

9 or 10 in. so that thin flat sheet wings can be used (approxi¬

Fig. 16
mately 1/16 in. thick) with no attempt to shape them into an
aerofoil section. Rubber models may be larger, in which case
The performance of such models can be quite amazing, with
some attempt may be made both to stiffen and improve the
flights of 30 to 45 seconds obtained regularly with correct
efficiency of the wing by adding curvature or camber. This trim and launching technique—a sort of a “cricket-ball”
throw upwards—and much longer durations possible under
favourable conditions. They are an excellent type for the
younger enthusiast to concentrate on since they are relatively
inexpensive and easy to build, and also provide scope for
trying out ideas in design layout, etc. Also they can be flown
from comparatively small flying fields, such as a school sports
ground or local park.
Above a wingspan of about 18 to 20 in. performance will
suffer because of the weight of solid balsa construction for
wings. However, for flying for fun, solid construction can be
retained for gliders up to about 36 in. span. Too big for hand
launching to any reasonable height, such models are towed up
Fig. 15 to a height by tow-line or a simple catapult (Chapter 4).

W
 AEROMODELLING ALL-BALSA MODELS
32 33
Design and constructional details normally follow on the lines is desirable. The small- to medium-size sports type control-
shown in Fig. 16, with the emphasis on simple assembly allied line model is a typical example (Fig. 17).
to good aerodynamic shapes. Here wings are cut from solid balsa sheet, shaped to a suit¬
The success of such a model, apart from being basically able aerofoil section just like a chuck glider wing. A light-
sound in design proportions, depends very largely on correct medium or medium grade of balsa would normally be chosen
choice of balsa—and so the type represents an excellent exer¬ in order to provide adequate strength with a relatively thin
cise in balsa selection. The wing panels represent the greatest section. Tail parts are again cut from balsa sheet, this time
wood volume, so the grade chosen should be very light in order thinner and only lightly shaped to an equivalent aerofoil
section by rounding the leading edge and tapering the trailing
edge. The fuselage, however, is built up as a hollow box from

FORMERS STRIP BALSA

Fig. 18

sheet balsa sides and balsa blocks for the top and bottom. One
or two sheet balsa formers are incorporated to shape the sides,
together with a plywood front former which both carries and
strengthens the motor bearers cemented to the sides, and forms
an attachment point for the bent wire undercarriage. From
this basic built-up box form, the fuselage is finally carved and
to keep weight to a minimum. The “cut” is not so important sanded down to a suitable shape before attaching the wings
since there is sufficient depth of section to provide strength and tail unit to complete the model.
against bending. Tailplane and fin parts should again be cut Sheet fuselage construction is also popular for free-flight
from very light sheet, this time quarter-grain for stiffness. power models. Here the sides are cut from fairly thin balsa
The fuselage boom should be of hard or very hard balsa for sheet of light or light-medium grade and then generally
maximum strength, with the other balsa blocks forming the braced on the inside with strip balsa. The two sides are then
“pod” section selected from the lightest balsa available. joined by sheet balsa formers and cross braces or strips and
For more advanced types of models sheet and block balsa finally covered with sheet balsa on the top and bottom after
construction may again be applied, but in somewhat modified internal details and fittings have been completed (Fig. 18).
form. This is particularly the case where total model weight Again where weight is not critical, and something more
is not too important, and simplicity and rapidity of construction attractive than a box shape is required, the fuselage top may
3
 ALL-BALSA MODELS 35
34 AEROMODELLING

penalty, by using thin balsa sheet of very light grade, overall

be covered with a light balsa block instead of sheet, carved to
a rounded decking. This form of construction is often favoured weight need not be prohibitive and strength in bending is
largely provided by the curvature or camber of the section.
for the larger low-wing radio-control models in order to arrive
Such wings, however, are rather prone to flutter and can
at a semi-scale appearance whilst retaining the simplicity of
prove troublesome in this respect when the model is being tow-
built-up box fuselage construction.
Solid balsa sheet construction is standard practice for launched. They are not widely favoured as a consequence,
practically all control-line models, regardless of size and type but some excellent contest models have been produced em-
(although larger stunt models may employ built-up tissue or
nylon-covered tailplanes to reduce weight). Sheet construction

is alsc widely used for power model fins and for glider fins, but
not on fins for larger rubber models where weight-saving
again is important (and fin areas are relatively large in any Fig. 20
case). Solid sheet wings are also ruled out for all type of flying
models, both free-flight and control-line, where the span bodying this form of wing construction. It is not recommended
exceeds about 20 in. (with certain exceptions in the case of for the average builder since he will almost invariably produce
sports models). a wing which is too heavy.
Normal construction in such cases is a built-up balsa frame A form of all-balsa construction for wings which is becoming
which is subsequently covered with tissue (small or medium- popular for small to medium-size radio-control models is,
size models) or nylon (larger models only). In many cases, basically, a more or less conventional wing framework sheet
however, wings may be partially sheet-covered, both to im¬ covered top and bottom. However, instead of building the
prove the efficiency of the wing by maintaining the aerofoil wing frame and then covering it with sheet, the wing is built
section over the front part of the wing, and also to add stiffness directly on to the bottom sheet, thus simplifying assembly and
to the wing against bending. In some cases sheet covering making for greater accuracy. The resulting wing is heavier
may extend over the whole wing (e.g. larger control-line than a conventional built-up tissue-covered wing, but not all
models and radio-control models). that much heavier, with suitable choice of balsa grade. It is
Again there are exceptions. Some contest gliders employing considerably stronger and far more resistant to puncturing in
very thin wing sections may employ all-sheet construction on sizes which would normally be only tissue-covered (not nylon-
the lines shown in Fig. 19. Although this implies a weight covered) anyway. However, for even better durability—and
36 AEROMODELLING

ease in getting a good finish—it is usually better to tissue-

cover such an all-balsa wing. The resulting increase in weight
is negligible compared with the benefits.
All-balsa wing construction of this type is applicable to
radio-control and sports type power models of between 20 and
chapter 4
48 in. wingspan as a typical range of sizes where it offers
advantages without imposing an undue weight penalty. The
TOW-LINE GLIDERS
model shown in Fig. 20 is an outstanding example.
Tow-line gliders are the least demanding of all flying models
and, of course, cost nothing to operate. As a consequence they
are immensely popular and are built in a wide variety of types
and sizes from about 2 ft. wingspan up to 10 or 12 ft. wingspan
or more. As a general rule performance improves with size
(this applying both to tow-line stability and flight performance),
although a large model presents quite a problem as regards
transport to and from the flying field, as well as costing more

Fig. 21

initially and taking longer to build. A good average size,

therefore, is between 4 ft. and 6 ft. wingspan, which range also
embraces the main competition class (for A2 gliders, see
Chapter 16).
Scale-type sailplanes are in general disappointing, for their
37
 TOW-LINE GLIDERS
38 AEROMODELLING 39
basic layout is unsuited to good model flying requirements. consistent with satisfactory stability. Since tailplane “power,”
Basically their wings have too narrow a chord to be efficient as regards its stabilizing or corrective effect, is equal to its
area multiplied by the distance of the tailplane from the centre
scaled down to model sizes, and are difficult to make strong
of gravity or balance point of the whole model (tail moment
enough and light enough with model construction. Also their
arm), “power” and thus stability can be maintained with a
fuselages are too short and tail areas too small to suit model
smaller tailplane area by increasing its moment arm. This
stability requirements. All the best model gliders, therefore,
are normally original designs and top performance is realized leads to the typical long-fuselage glider design, so different
from that of a full-size sailplane. To reduce the surface area
and thus the drag of a long fuselage, the cross-sectional area
is kept down to a minimum— hence the “stick” appearance. In

order to balance the weight of this long fuselage its length may
also be extended forwards a considerable distance, so that a
by the highly specialized contest designs employing what are
smaller nose weight is needed for balance.
virtually “stick” fuselages and typically model-type aerofoil
Such considerations as these can lead to extremes in design,
sections (Fig. 21). These are representative of the A2 class.
but the general pattern for A2 gliders is as shown in Fig. 21.
Other original designs may employ more orthodox propor¬
Wing-aspect ratio (ratio of span to chord) is higher than with
tions, conforming to typical values summarized in Fig. 22.
other types of models, to increase wing efficiency, but limited
Fuselage shape is relatively unimportant, giving considerable
by the fact that wing chords under about 3 J in. are very
scope for individual preference in design ranging from func¬
inefficient, as well as by structural considerations.
tional (e.g. a basic “stick” fuselage of minimum cross section)
The more orthodox gliders may not have the same still-air
to semi-scale layouts, etc. Some common variations are shown
in Fig. 23. performance, but can be equally satisfying for sports flying,
particularly as longer tow-line lengths can be used to get
Gliders to the A2 contest specification are limited in total
greater height from a launch and thus longer flights. In suitable
area (i.e. combined area of the wings and tailplane). In order to
weather flight duration can be enhanced by thermal currents,
get the most efficient disposition of area designers usually con¬
when a well-trimmed model may soar to a considerable height
centrate on putting as much of the total area as possible into
and quite commonly fly away out of sight, unless fitted with a
the wings, reducing tailplane area to the minimum possible
40 AEROMODELLING TOW-LINE GLIDERS 41

device to bring it down after a predetermined time (see veer off to one side and sideslip into the ground for a damaging
Chapter 11). This applies to all free-flight models, but tow- crash landing. Alternatively, it may veer from side to side on
line gliders in particular rely on thermal currents to prolong the tow-line, eventually slipping off prematurely.
flight duration. The first requirement is that the model should be trimmed
An essential feature of any tow-line glider is that it should be for straight flight, as any tendency to turn will be exaggerated
stable enough to tow up straight for launching. The method by the increased speed when towing. The best tow-hook posi¬
of launching is to lay out a suitable length of tow-line (usually tion can only be determined by experience with a particular
thread or nylon fishing line) in a downwind direction. The end design. It is normally on a line about 40 degrees from the
of the line is tied to a wire loop or ring which hooks on to a balance point, in the centre of the bottom of the fuselage
tow-hook in the bottom of the glider fuselage. A streamer of (Fig. 25). If too far back, the model will tend to pull off to
silk, nylon, or similar light cloth is also tied to the line near the
ring (Fig. 24).

Fig. 25

one side or the other. If too far forward, the model may veer
from side to side. The optimum position is affected by changes
in balance point or trim, and may need to be farther forward
for windy weather than calm weather. Often, therefore, alter¬
With a helper holding the model, the launcher then runs native tow-hook positions are provided on a glider, or pro¬
forward into the wind—and at the same time the helper re¬ vision made to adjust the tow-hook position.
leases the model in a slightly nose-up attitude with the wings Even with a correct tow-hook position a glider will not tow
level. The launcher then continues to tow the model forwards up straight if the wings or tail are warped or out of line. These
and upwards, in a similar manner to flying a kite. The speed are rigging faults which must be corrected, if satisfactory tow¬
at which the launcher moves forward depends on the speed launching is to be achieved. Also there may be a basic fault in
of the wind and the size and weight of the model. The model the design which makes it unstable under tow. The only cure
must not be towed too fast, otherwise it will overstrain and then is to correct the design fault—having first established
possibly break the wings. At the same time it must be towed that the lack of tow-line stability is not due to warps—such as
fast enough through the air to climb to the full height of the by adjusting the fin area. A simple increase in fin area, how¬
tow-line, when a slackening of the tow-line will allow it to ever, is not an automatic cure. Many smaller gliders suffer
slip the ring off the tow-hook and commence free flight. from “marginal” tow-line stability so that the slightest in¬
If the tow-hook position is not correct, or the model is badly accuracy in construction and trimming makes them difficult or
designed or trimmed, it will not tow straight. Instead it may impossible to tow properly. Larger models of good design
42 AEROMODELLING
TOW-LINE GLIDERS 43
should not suffer from this as an inherent fault, although the straight and locked by a wedge for tow-launching. This wedge
degree of tow-line stability in different designs may vary is attached to the tow-line by a short length of thread.
considerably. With the first method, all the time the tow-line is pulling
It will be appreciated that having trimmed a model glider forward on the tow-hook it will also pull the rudder-line
for straight flight to get a straight tow, it will then also fly in a forwards, holding the rudder straight. Once the tow-line falls
straight line when released at the top of its launch (which, clear, the rudder is pulled over into the turn position. With the
incidentally, with a good model and the tow-hook far enough second method the rudder is locked in a straight position until
the tow-line falls off. In releasing from the model it also pulls

Fig. 27

out the wedge holding the lever locking the rudder in the
Fig. 26 straight position, again allowing the rudder to move to the
turn position. There are other variations on these themes, but
back, should be almost overhead of the launcher so that the
all work in a similar manner.
model has climbed to a height equal to the full length of the
Tow-launching is essentially a two-person affair—one to
tow-line). There are two basic methods by which a straight
hold the model and one to move forwards with the tow-line.
tow can be followed by circling free flight.
To assist in recovery of the line ready for the next flight it is
(i) Putting the tow-hook on one side of the fuselage and
usual to tie the free end of the tow-line to a simple winch. Thus,
trimming for a straight tow by offsetting the rudder to com¬
when the model is released, the line can be reeled in rapidly.
pensate for the offset pull. For single-handed launching a catapult may be used, although
(ii) Using an auto-rudder device which pulls the rudder
this is rather more restricted in the length of line which can
straight for the tow launch but allows it to move to a position
be employed, and in the height gained by the launch.
for turning flight when the model commences free flight.
The basis of a catapult is a length of rubber strip tied to a
Method (i) is not a satisfactory solution, for the offset effect
stake which is pushed into the ground, the other end of the
will vary with different wind strengths and towing speeds. An rubber being tied to a length of tow-line. The best proportion
auto-rudder device is much more reliable and is quite simple
of rubber-length to line-length is about one to four to one to
to rig. The basis of all such devices is that the rudder is hinged three—i.e. a 100 ft. length of tow-line will need about 25 to
and pulled in one direction by a light rubber band attached to
33 ft. of rubber tied to it. It is important that the catapult
a rudder horn (Fig. 26). An adjustable stop is fitted to limit the
rubber is not too strong. Rubber strip f in. wide is quite ade¬
movement to that required for steady circling flight. A line is
quate for launching medium-size models via a 100-ft. tow-line.
attached to the other side of the rudder horn and taken either Larger models may need 3/16 in. or possibly J in. strip. It is
to a ring which slips over the tow-hook in front of the tow-line better to have the catapult rubber too weak rather than too
ring; or to a lever which is moved fowards to pull the rudder strong as if it is too strong it will be difficult or even impossible
 TOW-LINE GLIDERS
44 AEROMODELLING 45

to achieve a satisfactory launch. A weak catapult, on the other to get initial height from which the glider can start free flying
hand, may well produce successful launches with a model which and thus turn in a satisfying flight duration. Scope for “high
has insufficient tow-line stability for satisfactory running start” launching is also provided by hills, where the model can
launches with a conventional tow-line. be hand-launched. With a suitable terrain, and the right type
For competition work the length of tow-line is limited of model, it is possible to make a model slope-soar by launching
(Chapter 16) and a running launch is specified. Flying for fun, outwards from the slope into the wind direction. This does,
however, either type of launch can be employed, and with a however, demand a model with good “weathercock” stability
running launch the length of tow-line can be selected to suit to continue heading into the wind, rather than circling back
the conditions, and the size of the flying field. Launching heights into the side of the hill.
of less than about ioo feet are seldom satisfactory, except on The best possibilities here are realized by fitting the glider
dead calm days, since air near the ground is always turbulent with radio control when it can be “tacked” up and down the
and models do not perform at their best under such conditions. windward side of the slope, maintaining or gaining height in
Longer lines giving launching heights of 200 to 300 ft. are far the upward deflected air in front of the slope. In this respect the
more satisfactory. Provided there is sufficient space available, model is duplicating the performance of full-size sailplanes and,
and a very light line is used, tow-line lengths of up to 500 ft. with the proper type of model and suitable conditions, very
or more can be used. With very long line-lengths, however, it long flight durations can be put up.
will become increasingly unlikely that the model can be towed Equally, of course, radio-control gliders can be tow-launched
up to the full height of the line and it may have to be launched from level ground, when the control available is largely directed
with considerable sag remaining. towards keeping them within bounds, as well as circling and
As regards towing technique, this is something which can taking advantage of any thermal lift which may be present.
only be mastered with practice. The first thing is to get the
model stable under tow, and then to practise technique.
Basically, the launcher should aim to tow the model at a
minimum speed consistent with it moving forwards and up¬
wards. Any excess speed is only overstressing the wings and
will tend to show up any discrepancies in trimming or rigging.
In windy weather it may even be necessary to move towards
the model, rather than tow it forwards, in order to keep its
speed through the air down. And if the model does get into
difficulties, the best way to save it is either to move rapidly
towards it to reduce the pull on the line or to release the model
immediately.
There are other variations on tow-launching, such as starting
with a long length of line and winching it in instead of running
with the line. This has some virtues for sport flying. Pulley
launching systems are also sometimes used with larger models,
but the complication is seldom justified and there is less control
over the model than with conventional tow-line launching.
The whole basis of tow-line launching is, of course, simply
 RUBBER-POWERED MODELS
47
To absorb the power of such a heavy rubber motor and give
a long motor run (up to two minutes duration) a large pro¬
peller is needed, the diameter in such cases approaching one-

CHAPTER 5
I half of the wingspan. A generous amount of dihedral on the
wings will be needed to counter the torque of such a large
propeller, and both tailplane and fin areas will need to be
RUBBER-POWERED MODELS
relatively large. The design may therefore end up as a true
“original,” looking very different from any full-size aeroplane
Where maximum performance is the aim the success of a
rubber-powered model depends primarily on (i) very low
structural weight and (ii) the use of a relatively large rubber
motor. Ultimate performance—ignoring thermal assistance—
is also related to the size of the model. The bigger the model,
in general, the better its potential still-air flight duration, al¬
though there is a definite upper limit at which the correspond¬
ing size of rubber motor becomes too large and too powerful
both to wind comfortably and to be accommodated in a con¬
ventional fuselage structure. This upper limit is represented by
a model size of about 300 sq. in. wing area. The best perfor¬
mance is usually given by a somewhat smaller model (around
200 to 220 sq. in. wing area), this size corresponding to the
Wakefield International specification, although the current
Wakefield specification restricts rubber weight (Chapter 16).
(Fig. 28). Usually, too, since the very large propeller would
This restriction on rubber weight was introduced because
represent a prohibitive drag penalty when the power ran out,
Wakefield models as originally developed had too high a per¬
the propeller blades are arranged to fold back as soon as the
formance, being readily capable of 4- to 5-minute flights
model commences gliding flight.
without thermal assistance and presenting problems as to the
With the change in the Wakefield formula this type of high-
space required for flying them.
performance rubber model has virtually disappeared, except
A minimum structural weight demands the use of built-up
for “open” contests. It does, however, represent about the
tissue-covered structures throughout, and particular attention
most efficient type of flying model aeroplane ever produced,
to the selection of balsa wood grades and “cut.” Wood sizes are with an absolute premium on skilled design and construction.
also reduced to a practical minimum so that the “maximum It can obviously be beaten for sheer performance by a power
performance” rubber model is somewhat flimsy and readily model where the engine can run just as long as you want the
damaged if maltreated. Rubber weight may account for one- model to climb, and because performance can be obtained far
half the total weight of the model—thus a 44 to 48 in. span more readily in this manner the power model has become far
high-performance rubber model may only weigh 8 ounces more popular.
complete, of which 4 ounces is accounted for by the rubber Restricting the rubber weight whilst retaining a minimum
motor and 1 ounce by re propeller, leaving only 3 ounces
model weight, as in the current Wakefield contest model
for the whole airframe.
formula, means that structural weight can be increased for a
46

Li
40 AEROMODELLING RUBBER-POWERED MODELS 49

given size, and thus the airframe can be made stronger. Per¬ proportions. Fitted with an undercarriage, however, a rubber
formance is also restricted as a consequence, and the type model will take off successfully from any flat and reasonably
becomes a highly specialized contest type with limited appeal, smooth surface, which itself is a “realistic” feature.
and all designs conforming to a more or less similar basic Constructional features of a typical “general purpose”
layout (Fig. 29).
For "sport” flying—as opposed to contest flying—a more TISSUE COVUUNS
RUD&E!
rugged model is desirable, involving an increase in the air¬ TAILPLANE ua

frame weight. In order to avoid an excessive total weight,

DIHEDRAL BRACE

_/5% WING AREA

4B-S0 SPAN ABOUT 30%
WING AREA REAR MOTOR FIX
GUSSETS' DOWEL

LONGERONS
S CHORD
POLYHEDRAL
WING

LONG TA/LPLANE
LARGE DIAMETER MOMENT ARM
FOLDING PROPELLER SPACERS -UNDERCARRIAGE LEG

NOSEBLOCK

\ BALANCE
RUBBER MOTOR LENGTH

rubber model of small to medium size are shown in Fig. 30.

rubber weight is decreased, again leading to some loss in per¬ This is basically a “duration” design aimed at giving a good
formance. Also, since the rubber motor is smaller the propeller flight performance, but using a freewheeling rather than a
can be made smaller in diameter—with a diameter down to folding propeller, and retaining an undercarriage. Other types
one-third of the wingspan, for example. will normally follow a similar form of construction but may
A typical .model of this type can still have a very good per¬ trend more towards “original” or “semi-scale” features. In the
formance, and may even be given a semi-scale appearance by former case a larger diameter folding propeller would be used
the incorporation of a cabin in the fuselage shape. To avoid the (normally a single-blade propeller with counterweight for
complication of a folding propeller the propeller can be made simplicity), polyhedral used on the wing instead of dihedral,
to freewheel or “windmill” at the end of the power run, and the undercarriage omitted, and the fuselage shape reduced in
so reduce its drag in this manner. If the model is designed cross-section (probably mounting the wing on a pylon). For a
primarily for performance, then the undercarriage may be semi-scale variant the fuselage shape would be made more
omitted entirely (as on virtually all modern contest models). “realistic,” even if this meant an increase in bulk and weight.
If it is a semi-scale type an undercarriage is fitted, although the More attractive outline shapes may also be used for the tail-
length of leg necessary to provide ground clearance for the plane and fin, and possibly the wing tips. The model design
propeller will represent a considerable departure from “scale” would, in fact, be “styled” around similar basic proportions.
50 AEROMODELLING

A similar form of construction is used again in the rubber-

powered flying scale model. A good compromise between scale
realism and model requirements is, however, difficult to
achieve. The necessary increase in tail areas may not be obvious
if the same outline shapes are retained, but there is no escape
from the fact that the wing dihedral must be increased from the

C/D C/D
m m
a <N

CO C/D
«0N<

Fig. 31

scale value, and the undercarriage length increased to accom¬

modate a suitable size of propeller. If the propeller diameter is
reduced (to preserve a more realistic undercarriage length)
performance will automatically be reduced since the matching
size of rubber motor will also be reduced. Thus although the
rubber-powered model is attractive from the point of view of d
c
basic simplicity and low cost, it is not particularly suited to .5
£ O.173
to CO
;> co ACO
flying scale designs. Power models are much better in this
respect, although similar stability problems are involved.
Typical design proportions for an original rubber model are z
w O
shown in Fig. 31. Corresponding dimensions for typical model >< p
sizes can be worked out from the proportions summarized in H
Table VI. Actual shapes are largely immaterial, provided they p
 RUBBER-POWERED MODELS 53

of rubber strip required is equal to the made-up motor length

multiplied by the number of strands.
The number of strands required, and the size of the strip
section, is determined by the size and weight of the model and
by the propeller diameter and pitch. Propeller pitch refers to the
angle of the blades, or, more correctly, the theoretical distance
the propeller would advance in one revolution if screwed into a
solid medium (like a screw being screwed into wood). The
power of a large rubber motor may be absorbed by increasing
propeller diameter or pitch. Thus, for example, a 44 in. span
model may employ an 18 in. diameter propeller with a high
pitch (1 - 75 to 2 times the diameter); or, say, a 24 in. diameter
propeller with a medium to low pitch (1-5 to 1 times the
diameter). Such factors can only be worked out by experience,
but for any given design a propeller size is always specified,
together with the required power (number of strands in the
motor). Table VII can also be used as a general guide.
No recommendations of this kind can be exact. If the model
is heavier than expected it may need a greater number of
strands in the motor. If it works out lighter it may be possible

TABLE VII. TYPICAL RUBBER MOTOR SIZES

Duration Model 30 in. 36 in. 40 in. 44 in. 48 in.

to
00
Prop. Dia. (in.) 12-14 l6 18-193

1
■
No. of Strands
of £X24 Strip1 8 10-12 12-14 14 l6

Semi-scale
Model Span 18-24 in. 33-36 in.

Prop Dia. (in.) 6-8 8-9 9-m IO-II 11-12

No. of Strands

CO
of i X 24 Strip

t-H
0
4-6 6-8 10-12

1
1 Or equivalent in other sections.
* 20-22 in. dia. with low pitch.
* 22-24 in. dia. with low pitch.
54 AEROMODELLING RUBBER-POWERED MODELS
55
to reduce the number of strands (using the same weight of A simple and very effective method of accommodating this
motor) or increase the propeller diameter, either of which slack is by “cording” the motor. To do this the motor is laid out
could improve performance. Also if you “guesstimate” wrongly in two “legs,” each having half the required number of strands
in an original design you may have to adjust the number of (see Fig. 33). The middle point (C) is lightly bound to a suit¬
strands in a motor, this being simpler than carving a new able marker (such as a short length of dowel) with a rubber
propeller. band, the complete motor is opened up, end B held and a
Made-up motor length is rather arbitrary. Using a maximum winder attached to end A to wind on about 100 to 200 turns
(depending on the actual length of the motor—experience will
indicate the actual number of turns required). Ends A and B
are then brought together again (putting end B on the winder
hook), and the motor pulled out fairly taut by holding at C.
Release the winder and allow the motor to unwind, when it
will twist up into a “cord” much shorter in length than the
original motor length. Each end of the corded motor C and
A and B together is then bound with a rubber band to hold it
in place, when the motor can be inserted in the fuselage. It can
then be wound up fully in the normal way. On unwinding it
will always revert to its shortened “corded” form and thus
never fall slack in the fuselage.
rubber weight for best duration performance, length will If it is still slack then more “cording” turns are required. If
follow automatically from the number of strands required. For it is too tight between anchorages when unwound, then too
all other models a good general rule is that made-up motor many “cording” turns have been used. This is not necessarily
length should be approximately the same as the wingspan, a bad thing, except that the number of cording turns put on
provided the model has orthodox proportions. This will ensure reduces by a similar amount the number of turns you can wind
a reasonable size of rubber motor. It can, of course, be shorter, up the motor. Thus to get maximum number of winding turns
but this will detract from performance. It can also be longer, from a rubber motor use a minimum number of “cording”
but this will also increase rubber weight and thus model weight, turns—just enough to take up the slack when unwound.
and call for extra strands, again increasing weight. The number of turns which aero-strip will take can be esti¬
A satisfactory motor length, therefore, is invariably longer mated from Table VIII. This gives maximum turns for a
than the distance between the propeller shaft hook and the complete range of number of strands in all standard strip
rear rubber anchorage in the fuselage, unless the fuselage is sizes and allows about 10 per cent safety margin. That is, the
deliberately made long enough to accommodate this length. table figures are lower than the number of turns to break the
This is, in fact, done on some high performance designs, and motor, but they refer, of course, to good quality strip in good
particularly on contest designs with a “limited rubber” rule. condition and properly lubricated and broken-in.
With the average model, however, the motor length is always Lubrication is important both for achieving maximum turns
longer than the distance between anchorage, which means that on a rubber motor and for preserving its life. An unlubricated
when the motor is unwound it will lie unevenly on the bottom rubber will chafe on being wound, and readily break up. The
of the fuselage and almost certainly upset the balance for only lubricants which should be used are special rubber lubri¬
gliding flight. cant (based on soft soap and glycerine mixtures) or castor oil—
56 AEROMODELLING

never ordinary lubricating oil or grease which will only attack

and rapidly destroy the rubber. Lubricant should be well
rubbed into the motor immediately after making up into the
first large loop (not before, since it is difficult to get knots to
hold in lubricated rubber without binding with wool). Use

TABLE VIII.
MAXIMUM SAFE TURNS FOR RUBBER MOTORS
(No. of turns per inch of motor length)

Number Rubber Strip Section

OF
Strands iX24 ix3° iirX24 -&X30 ix 30

2 60 63 66 72 f9°
4 46 47 49 5i 63
6 36 39 4i 44 51
8 30 33 35 37 44
10 26 29 3i 33 38
12 24 28 29 3i 36
H 22 25 27 29 33
16 20 25 26 27 30
18 ^ —
24 24 27 30
20 — — 23 26 29
22 — — 21 25 28
24 — — — 24 26

PROPELLER BLOCK DIMENSIONS

enough lubricant to wet all the rubber thoroughly and then
DIAMETER 6-8" 10" 12" 14" 16" \ IB"
wipe off any surplus. Relubricate whenever the rubber appears 1 W /V I'l2" /V /%- 2" 2"
T /"
to be drying out. J/4" I'l4’ il/e" i'ii" !J/4"

A new rubber motor will break at about two-thirds or less of

Fig. 34. Carving a Propeller
its normal maximum turns if you attempt to wind it right up This is one of the trickiest operations in constructing a
first time. It must, therefore, be broken-in carefully, which rubber-powered model. The basic steps involved are—
(i) Mark out and cut a block of the correct proportions.
means winding up to something less than half maximum turns ,
(ii) Mark out the block and cut to a blank shape.
and letting unwind; repeating with about 10 per cent additional (iii) Carve the blank edge-to-edge to produce a correct
turns; and so on until you have reached 80 to 90 per cent propeller form.
maximum turns. This will normally involve some five separate
windings and unwindings, after which the motor can be in¬
stalled in the model ready for use.
58 AEROMODELLING

There is also an art in winding a rubber motor. Apart from

convenience, a winder is essential to put on maximum turns
since the motor must be stretched as it is wound-—known as
stretch winding. At the commencement of winding the motor
need be only lightly stretched, but the operator should then chapter 6
move outwards from the motor until the motor is stretched to
at least three times its original (made-up) length. Hold this POWER MODELS
position until one half to two-thirds the number of turns re¬
quired have been wound on, then advance towards the model With such a wide choice of miniature internal combustion
slowly as the remaining turns are applied. This will not only engine sizes available a power model becomes a practical pro¬
enable more turns to be put on without over-stretching the position in sizes from a little more than i ft. wingspan upwards,
motor but will also result in more even winding. with motor run limited only by the size of the fuel tank. There
is seldom any question of not having enough power; usually the
reverse is true, so the problem of building down to minimum
weights for maximum performance does not arise. As a conse¬
quence, power models are generally more robust than either
gliders or rubber models, and can also make use of simplified
construction techniques, as described in Chapter 2.
Small-size power models do, however, have their limitations,
being mainly suited for calm-weather flying. They are also
generally unsuited for competition work where motor run is
restricted to a matter of 10 seconds or so, and flight duration
depends largely on efficient model design capable of achieving
a very rapid rate of climb on the power available, followed by a
prolonged glide. Engine efficiency, too, increases with engine
size, again leading to better climb performances. On the other
hand the small model, because of its light weight, can often
take a lot of rough treatment in the matter of crash landings
without damage whereas a larger model might suffer serious
damage under similar circumstances. Small-power models
therefore, have a particular appeal for “sports” flying in sizes
up to about 40 in. span. Competition model sizes are usually
based around the recognized maximum engine sizes in the
competition classes (typically 1.5 c.c. and 2-5 c.c., the latter
being the International maximum size).
Free-flight power models fall into three definite groups—
original designs based on achieving maximum duration per¬
formance; semi-scale models; and scale models. The groups
are quite distinct, although there is not all that difference
59
60 AEROMODELLING POWER MODELS 6l

between semi-scale and scale as regards basic layout, matching and tends to give all such original designs a very similar appear¬
engine sizes, and performance. The duration models on the ance. Main differences, in fact, are usually in the structure.
other hand are an entirely separate type and completely un¬ Other configurations are not excluded completely, but are
rivalled as regards sheer performance potential. invariably trickier to trim and need intensive development to
The chief requirement in this latter respect is a design layout be brought up to comparable contest performance standard.
stable enough to be able to accommodate the considerable Virtually every possible alternative configuration has been
power available from the matching size of engine—power tried, but the pylon model still remains the normal and
which may result in a climb performance of several thousand safest choice.
feet per minute (although, as noted, engine run is restricted to It will also be appreciated that no attempt to add semi¬
a fraction of a minute for competition work). The design layout scale feature to such a layout can provide more than a gesture
which has proved most capable of controlling such power is the towards “realism”; and any attempt to modify the proportions
in order to achieve realism can only drastically detract from
the stability margin provided by the original layout. In prac¬
tice, this means that a semi-scale model although retaining
perhaps a similar wing and tail cannot accommodate the same
engine power without becoming unstable. It cannot, therefore,
begin to compete as regards contest performance. Thus the
semi-scale model is basically limited to non-contest flying; or
in further developed forms for radio control (see Chapter 14).
It is, however, far more popular in numbers, and is also far
less exacting to trim. Although the pylon layout provides
additional stability, it is still a tricky model to trim with a
powerful engine and requires considerable experience to handle
successfully. On the other hand, for a modeller who is seeking
“duration” performance rather than “realism,” the duration-
type model fitted with a smaller and less powerful engine can
be a safe and easy model to fly.
Typical proportions for a semi-scale power model are shown
pylon configuration, as shown in Fig. 35. Note that in addition in Fig. 36. A high-wing cabin layout is the most popular choice
to being mounted well above the fuselage and right forward since this basic configuration has a reasonable degree of sta¬
(just behind the engine), polyhedral is also standard, with bility to start with. Shoulder-wing models can be a little more
generous dihedral angles on the outboard panels. tricky to design and trim, and low-wing models even trickier.
There are variations in proportions, such as in fuselage The latter do not, in fact, normally make good free-flight
length, fuselage shape, and mounting the engine in line with models, although with careful development and attention to
the wing in a nacelle rather than on the fuselage, but the specific requirements (such as an increase in dihedral angle
basic pylon layout remains the standard. The modern trend compared with high-wing models and readjustment of fin area
has been towards reducing the cross-section of the fuselage to to balance) they can be suitable for sports models. Because the
“stick” proportions (normally hollow sheeted box construc¬ stability margin is less than in a high-wing monoplane, it is
tion), which emphasizes even more the pylon configuration also desirable to reduce the power, that is, to use a smaller
02 AEROMODELLING POWER MODELS
03

engine in the same size of model, compared with a typical Increasing the dihedral above the true scale value will alter
high-wing monoplane, or make the model larger for the same the appearance, and is in any case a departure from true scale.
size of engine. A secondary effect is that increasing the dihedral must also be
Somewhat similar considerations apply to biplane models. followed by increasing the fin area, to balance the additional
Stability is generally better than that of a low-wing model, but side area of the wings. The fin size will probably be too small
the shorter wingspan provides less control for the torque of the for model stability requirements to start with, so it may need
increasing by a substantial amount to arrive at a stable model
layout. The tailplane will also almost certainly be too small for

MODERATE
ENGINE SIZE

Fig. 36
model stability—it needs to be at least one-fifth of the wing
motor. To compensate for this the model must not be under¬ area, and even bigger to be on the safe side.
powered rather than overpowered (which would only aggra¬ On this basis it is a wonder that flying-scale models will fly
vate the rolling action of torque). at all! There is also one further difficulty. Relatively few full-
With a scale model the difficulties of obtaining stable flight size aeroplanes have simple box-shaped fuselages and dupli¬
are usually exaggerated, unless the designer is prepared to cating a rounded fuselage will add structural weight. This
sacrifice true scale outline and appearance in order to arrive at means more power is required to fly the model, and it will fly
what approximates fairly closely to semi-scale proportions. faster as a consequence. That in itself is not necessarily im¬
The two chief difficulties arise with regard to wing dihedral and portant, except that the faster the model flies the faster it will
tail areas. For satisfactory automatic stability a high-wing land, and thus the more liable it is to damage itself in a bad
free-flight model needs a minimum of about 8 degrees of landing. What is important, however, is that the faster flying
dihedral, and 10 degrees is better still. A shoulder-wing model speed will make stability requirements more important.
will need more dihedral, and a low-wing model more still (as There are answers to most of these problems, if not com¬
much as 15 degrees). Full-size aeroplane wings generally have plete ones. The first, and most important, is to choose a full-
very small dihedral angles, perhaps only a matter of a degree size prototype which has proportions fairly close to semi-scale
or so on high wing layouts. model requirements to start with, preferably with a simple
64 AEROMODELLING

fuselage shape which can be duplicated in a simple model

structure without adding excess weight. Above all, avoid the
sleek, streamlined low-wing aeroplanes which may look attrac¬
tive but will be hard to build, and even harder to get to fly
without extensive modifications of proportions. To start with,
some have little stability as full-size aeroplanes. The same

TABLE IX. TYPICAL SIZES OF “DURATION” POWER MODELS

outlines can be quite hopeless when applied to a free-flight
model where automatic stability is essential for satisfactory
flight.
Having settled on a suitable prototype, it is then a matter of
compromise on such matters as dihedral and tail areas. To play
safe, you increase these values. To retain true or very near
scale outlines you retain the original proportions, or exag¬
gerate them only slightly (so that appearance is largely un¬
changed). The model will have less stability than a normal
free-flight model as a consequence, but it may have enough to
be flown under good conditions, that is in calm weather. That
is another basic rule. The lower the margin of stability in the
model design, the more important it is to restrict flying to calm
weather when there is less chance of a gust upsetting it and
showing up the deficiencies in stability.
Another important thing is to avoid all excess weight, not
just in the structure but in doping and finishing. Colour dopes,
generously applied, may result in an excellent appearance, but
can add a lot of weight. Model size and “permissible” maxi¬
mum weight are closely related, and should follow closely
figures typical for semi-scale models (Table X). It is then
possible to get a satisfactory flying performance with moderate
to low engine power, and with the flight much safer as a
consequence.
An alternative solution to the lack of “model” stability
inherent in full-size aeroplane outlines is to provide the model
with a form of “automatic pilot” to apply correction automati¬
cally when needed. Devices of this type are usually based on a
simple pendulum mounted in the fuselage and connected to the
tail surface controls. If the pendulum is limited to swing in a

Loading
fore and aft direction, control can only be applied to the ele¬
vators (Fig. 38). If freely pivoted, the pendulum can be linked
to both elevators and rudder. Alternatively, the rudder only
66 AEROMODELLING POWER MODELS 67

may be controlled by a simple pendulum mounted in a fore portant, and in the case of the rudder a movement of two or
and aft direction, the rudder being the most powerful control three degrees only is the maximum which should be used,
in any case. otherwise there is a distinct possibility of the pendulum “lock¬
The principle involved is extremely simple. Consider pendu¬ ing” the rudder with the model entering and remaining in
lum rudder control first. If the model starts to slip or bank to a spiral dive.
one side the pendulum also swings to that same side applying Pendulum control may also be extended to ailerons as well
opposite rudder movement to correct the departure from the as rudder and elevators, although with this more complex
original flight path. Similarly with pendulum elevator controls. hook-up “auto-pilot” action is likely to be much more erratic.
Such a system is not to be recommended until one has tried
several models with simpler pendulum controls and appreciate
their workings, and limitations. Pendulum controls are really
another “compromise” solution—not a complete or even
partially complete answer to stability problems with flying
scale models. The only corpjjete answer is full control of the
model all the time it is in fl^ht, which can be obtained only
with control-line flying, or multi-channel radio control.

If the model starts to climb excessively the pendulum swings

backwards, applying down elevator to correct. If the model
starts to dive the pendulum swings the other way to apply
corrective up elevator.
In actual fact no such simple device as a pendulum can
provide foolproof “automatic pilot” control. The pendulum
will be affected by acceleration of the model as well as changes in
altitude and under certain conditions can apply “corrective”
control movements in the opposite way to that required. Never¬
theless, in practice pendulum controls can give quite successful
results and make it possible to achieve reasonably stable flight
with models employing true-scale outline proportions. The
flight is unlikely to be perfectly steady, but with a suitable
design and position of pendulum and limited control movement,
the delay in applying correction at times, and the tendency to
overcorrect at others, may even make the flight more thrilling
and realistic—especially, say, in the case of a scale World War I
biplane. The necessity of limiting control movement is im-
 jet models 69
gases escaping when the charge is ignited. The end cap nor¬
mally seals on a gasket, which must be kept in good condition,
or replaced if damaged, to prevent leakage and loss of thrust
CHAPTER 7 during normal firing.
The charge—a special rocket fuel—is ignited by a fuse held
JET MODELS in contact with the end of the charge by a wire g^uze, Jhe free
end of the fuse being taken out through the jet orifice^fihat it
The only practical—and acceptable—-jet power unit so far can be ignited from the outside. The fuse materials aqfjally a
developed for free-flight models is the “Jetex” solid fuel rocket
available in a number of sizes giving from | ounce up to about
5 ounces thrust. Thrust duration is limited by the size of the
charge which can be loaded into a relatively small volume and,
typically, may range from about 5 to 10 seconds. The “Jetex”
is, therefore, a limited duratiow(Wwer unit and so for maximum
overall flight performance is best applied to “power-duration”
type layouts, but of much smaller size and lighter weight to
NORMAL CHARGE SHAPED CHARGE FUSE MAY BE ADDED
match the size and thrust of the “Jetex” unit. It can, of course,
equally well be applied to other types of free-flight models, Fig. 40
accepting that flight durations will be limited, and makes X.
chemiqal coating on a ver» thin wire and for maximum thrust
practical the small size jet scale model, as well as unorthodox
layouts which are difficult to fly with engine or Rubber-motor the wire'should be pulled free from the jet orifice as soon as the
power because of engine torque. The “Jetex?’ provides power charge is ignited, or loadeckimsuch a way that it will blow free
in the form of pure thrust, like a full-size engine, with a and so not partially block tl* jeft.
complete absence of torque. It is possible to obtain m<*e thrust from a standard “Jetex”
unit by “shaping” the char^ although this will also reduce
the duration of the thrust. “Shapmg” consists of cutting grooves
in the charge, as shown in Fig. 40, to promote more rapid
burning. Thrust may be roughly doubled (and duration
halved) by laying additional lengths of fuse in the* grooves so
that the whole of the charge is ignited along its length simul¬
taneously. Increasing thet burning rate of the charge in this
manner also increases thei heat developed, which may be
SEALING WASHER BODY OR CASING
enough to melt the aluminium casing! Special heat-resistant
Fig. 39 cases are used in such cases.
Thrust can also be increased by combining the “Jetex” with
A “Jetex” unit is shown in Fig. 39. The casing is of aluminium an augmenter tube (Fig. 41). The tube has a bell-mouthed shape
and fitted with an end cap held in place with springs or a at the front end into whion the rocket jet is directed, position
spring clip. This is a safety measure to allow the end cap to lift of the bell-mouth relative to the jet being fairly critical for best
off should the jet orifice be blocked in the end cap, preventing results. The overall length of the tube can be extended by
70 JET MODELS
AEROMODELLXNG 71
plugging on additional straight tube lengths to suit a particular
model layout.
Besides being a thrust booster, an augmenter tube is also
useful for accommodating the “Jetex” inside the fuselage of,
say, a scale jet model, enabling the “Jetex” to be placed at the

Fig. 41

balance point of the model, with the jet itself exhausting through
a tailpipe at the end of the fuselage. An alternative and
design and trimming is that although the thrust is fairly con¬
simpler method which is often used with smaller-scale jet
stant after the initial build up, the efficiency of the jet engine and
models is to mount the “Jetex” in an open “trough” formed in
thus the power developed increases with increasing speed. In
the bottom of the fuselage—Fig. 42. This is not, of course, true
practice, this means that under “Jetex” power a conventional
scale but has the virtue of making it very easy to remove the
free-flight model will tend to accelerate throughout most of the
“Jetex” for reloading.
power run, which will show up any faults in trimming or warps
in construction in the form of instability, e.g. a slight turn to
start with may end up as a steeply banked turn with the model
losing height under power; or develop a moderate climb into a
series of loops. Trimming a “Jetex” model, therefore, needs
care, and particular attention must be paid to seeing that the

TABLE XI. TYPICAL “JETEX” MODEL SIZES

Nominal Model Model Wing Total

Model proportions for stability are similar to those of other Jetex Thrust Span Chord Area Model
Unit (ounces) (inches) (inches) (sq. in.) Weight
free-flight models, except that with the absence of torque wing
dihedral can be kept to low values for scale models and tail (ounces)

surfaces can be smaller. Also, with a duration-type model,

Atom 35 f-£ IO-I2
exaggerated pylon mounting of the wing is no longer neces¬ 2| 25-30 M
50 1 18 3 50 f-i
sary. It is better, in fact, to locate the thrust-line fairly close to 100 i-ii 24-30 IOO-I5O
4 1-11
the wing, a typical layout being shown in Fig. 43, with sizes, PAA-Loader lf-2 24-36 4i 125-175 if-ai
etc., summarized in Table XI. Scorpion 4 32-44 5i 180-260 8-12
One peculiarity of “Jetex” propulsion which affects both
72 AEROMODELLING JET MODELS 73
“Jetex” unit is always mounted in exactly the same position in generated by the fan directed through the tube. It also tends to
its clip, as any slight change in position can disturb the align¬ be relatively inefficient as a method of producing thrust, unless
ment of the jet thrust and upset trim. Otherwise “Jetex” the fan is carefully made and closely fitted to the tube diameter.
flying is quite straightforward, with the cost a few pence per Tube shape is also important, some benefits being obtained by
flight for the charges. Although this may seem a lot, this cost increasing the diameter forward of the fan and decreasing it
is actually less in overall figures than the cost of engine- aft of the fan to connect to a smaller diameter tailpipe. Again,
powered flying. however, unless these changes in diameter are correctly pro¬
The “Jetex” also lends itself well to unorthodox models portioned the result will be a loss of thrust, rather than an

improvement. A parallel “straight through” tube may, in

fact, be the best proposition.
Fig. 44 The faster the fan can be driven, the better the effect. A fast
running speed will normally be ensured because the fan dia¬
which are not easy to power by other means, such as heli¬ meter is much smaller than a normal propeller fitting the same
copters, deltas, etc. (see Chapter 8). size of engine for free-flight thrust. However, the engine itself
The main limitation is that the size of even the largest may be limited as regards the ultimate speed it can reach with
“Jetex” unit restricts maximum model size to about 44 in. a fan type load. Glow motors tend to be better than diesels in
span, whereas the most popular “Jetex” sizes (’35 and ’50), this respect.
restrict sizes to about 18 in. span. Also, of course, all “Jetex” The fact that the engine is mounted inside the fuselage
units are capable of only a short power run, so flight durations (actually inside a tube in the fuselage) presents some problems,
are naturally limited. A “Jetex” unit is not suitable for a particularly as regards starting. The fan cannot be flipped over,
control-line model, for example, although it may be used for so starting must be done by a cord wrapped round a light rim-
“sprint” or “speed” models for tethered indoor flying (see type flywheel, or a pulley section incorporated on the fan.
Chapter 15). This means that the fuselage must have a hatch to give access
For larger “jet” models there is an alternative—ducted fan to the engine, and it is necessary to have this hatch very close
propulsion. Basically this comprises a hollow tube running fitting where it breaks the tube section in order to avoid air
throughout the length of the fuselage, in which is mounted a leakage and loss of thrust when closed.
conventional diesel or glow engine fitted with a multi-bladed Being a form of jet propulsion, the ducted fan also has
small-diameter fan or impeller (Fig. 44). This is not really true similar characteristics to “Jetex” in that its efficiency increases
jet propulsion, the model being propelled by the slipstream with speed. Such models, therefore, normally fly quite fast—
74 AEROMODELLING JET MODELS 75
if they cannot fly fast enough the thrust developed will be in¬ undercarriage retracted. They are also a very interesting type
sufficient to keep the model flying. They are thus not an easy for the serious modeller to develop.
type of model either to design or to build, but can be a most There is another type of jet engine produced, known as a
interesting and satisfying type of free-flight models when pulse jet and based on the same operating principle as the Vi
necessary details are worked out properly. Typical data re¬ or buzz-bomb power plant of World War II. This runs on
garding sizes, etc., are summarized in Tables XII and XIII as straight petrol, can develop several pounds thrust for less than
one pound in weight, and is capable of propelling models at
TABLE XII. TYPICAL DUCTED FAN MODEL DATA speeds up to and exceeding 150 m.p.h.! Because of its noise,
fire risk, and the potential hazard to the public and property
Model Span Wing Wing Fan of such fast-flying models, the use of pulse jets as a power unit
Engine Weight Area Loading R.P.M. Dia.
for free-flight models is banned in this country. It may, how¬
Size Orthodox Delta (ounces) (sq. in.) (oz. (in.)
sq. ft.) ever, be used for control-line flying, although again it is looked
on with little favour by the authorities, largely because the
o-5-o-8 c.c. 24 20 7-10 IOO-I10 9-io 10,000 3-3 i
I C.C. 28 12-16
noise may be heard over a distance of several miles and there
24 150-160 10-13 10,000 4
13,000 3i
is nothing that can be done to silence a pulse-jet.
1-5 c.c. 32 26 16-22 190-200 12-14 10,000 4l
13,000 4
2-5 c.c. 35 28 22-28 230-250 13-18 10,000 4i
15,000 4
0*29 cu. in. 44 30-40 300-330 14-20 10,000
32 5
15,000 4i
0-35 cu. in. 48 36 40-50 350-400 14-20 10,000 5
15,000 4i

TABLE XIII. TYPICAL FAN SIZES

0-5-0-8 cc. I C.C. 0*29 or

1-5 c.c. 2-5 c.c. 3'5 c.c.
0-049 Glow 0-09 Glow 0*35 Glow

Diameter 3 in- 3iin- 3f-4i in. 4-4i in- 4i in. 4f-5 in.
No. of Blades 9 12 12 12 12 12

a general guide to requirements. Suitable fan types are detailed

in Fig. 45.
Ducted-fan models also have certain advantages, apart from
being the only suitable form of power for larger free-flight jet
models. Since engine and fan are mounted inside the fuselage
they are protected against damage in hard landings, etc. Also
no undercarriage is needed, so that semi-scale and scale jet
models have all the realism of the full-size aircraft in flight with
 UNORTHODOX MODELS
77

CHAPTER 8

UNORTHODOX MODELS

Summarizing the main characteristics of free-flight models

described in earlier chapters: (i) all free-flight models need to

have “automatic stability” built into the design; (ii) original
designs usually fly better than semi-scale or scale models; and
(iii) light construction is essential for good flying performance.
These requirements lead to a general similarity between dif¬
ferent types of models. Models which embody quite different
shapes or proportions, or even different flying principles, are
generally classified as unorthodox models. Again they can be
completely original designs, or patterned on unorthodox full- with a long fuselage to accommodate a good length of rubber
size layouts. The latter are more usual since most of the motor, may be capable of flights of 1-3 minutes duration.
“flyable” shapes have been tried as full-size designs, and many Stability is a problem, however, with fixed angle rotors, with a
such shapes are quite well known. tendency for the model to fall to one side as it climbs and even
The full-size helicopter is a typical example. It is not an turn right over and dive into the ground. Altering the position
easy type to reproduce as a flying model, however, since it is of the fixed vanes on the fuselage can often cure this.
virtually impossible to duplicate the relatively complicated With engine power a semi-scale layout can be adopted in
mechanism of the rotor hub on which the stability and per¬ which the fuselage unit merely becomes an appendage sus¬
formance of a full-size helicopter largely depends. Model heli¬ pended from the rotor system. The rotor may be directly
copters thus merely employ the same basic principle of deriving driven by an engine-driven airscrew mounted on an arm on the
lift from a rotating rotor, and from then on are purely “ori¬ rotor system at right angles to the main rotor; or the rotors
ginal” designs. may be attached to the crankcase of the engine so that as the
In the case of a rubber-driven helicopter the design becomes
highly original or “unrealistic” since, in its basic form, it ROTOR ROTOR
consists essentially of a long stick-type fuselage with a large
propeller at one end, forming the rotor. To prevent the rubber
motor rotating the fuselage faster than the rotor and thus
wasting power, fins may be attached to the fuselage to restrict
its rate of spinning (Fig. 47). Alternatively, a second rotor
(large propeller) may be coupled to the other end of the rub¬
ber motor, rotating in the opposite direction to the first and
thus having opposite pitch. Such models, very lightly built and
76
78 AEROMODELLING UNORTHODOX MODELS 79

engine drives its own propeller it also rotates in the opposite TABLE XIV. DESIGN DATA FOR “JETEX” HELICOPTERS
direction under torque reaction and so drives the rotors. Both
schemes are shown in Fig. 47—and both can work quite well. Rotor Rotor Skew
The main requirement is a stable rotor system. A fixed angle Jetex Dia. Blade Hinge Blade
rotor is not stable—it produces the effect described above for (inches) Area Angle Incidence
(sq. inches)
rubber-powered helicopters. Thus stability has to be provided
by pivoting or hinging the rotor blades about the hub, allowing
them to alter their angle of attack as they rotate.
50
100
22
30-34
18
40
■a
H*
5°
7°
150 36 55 30° 7°
Scorpion 60 150 30° 7°

duces a lot of drag with very little lift. Autorotation produces

a reasonable amount of lift with fairly low drag. To fly success¬
fully, therefore, a model autogiro must have a freewheeling
rotor which will autorotate rather than windmill—something
which is fairly difficult to achieve. The power in this case is
provided by a conventional propeller on the front of the fuse¬
lage pulling the model forward (powered either by a rubber
motor or small engine).
Normally this requires that the rotor blades be set at a nega¬
tive angle (negative pitch) relative to the rotor shaft, with the
rotor shaft itself inclined backwards slightly. The thrustline of

The “Jetex” unit provides an even simpler method of TAILPLANE IS15X3 X #16
CEMENT ON F/NS AT ANGUS
powering a model helicopter rotor, using one or two “Jetex” GIVEN BY THE BRACES

motors mounted as shown in Fig. 48, and completely eliminates

torque problems. Again, hinged or pivoted rotors must be
used, otherwise the rotor system will not be capable of stable
flight. The other advantage of a hinged rotor system (for both 1/16 SO. RUBBER TIED 70 PIN - FIND BES T
ANCHOR POINT BY EXPERIMENT
“Jetex” and engine power) is that when the power runs out the
rotors will continue to autorotate in the same direction and thus
give a safe descent. ADDITIONAL
BRIDLE LINES ' l
Autorotation, where a rotor system is not power driven but WILL HELP IF
MODEL WILL NOT
,

KITE PROPERLY |
rotates under the action of an airstream is the principle of
KITESTRING
auto giro lift (or helicopter descent, when the power is shut off).
It is distinct from “windmilling” since the blades actually
advance into the direction of the airstream. Windmilling pro- Fig. 49
80 AEROMODELLING UNORTHODOX MODELS 8l

the pulling propeller must be angled downwards to a con¬ type of unorthodox model which attracts the serious experi¬
siderable degree. Engine torque will tend to roll the model, and menter.
this may have to be dampened by fitting stub wings with Practically all successful model ornithopters are rubber-
sharply dihedralled tips. Design requirements, in fact, are powered, largely because the flapping rate of the wings needs
quite tricky and have largely to be worked out on a “trial and to be kept fairly low (between one and two beats per second),
error” basis. This type of model can be made to fly quite well, which cannot be achieved by driving from a high-speed engine
however, and represents a real challenge to the modeller without heavy (and generally unreliable) gearing. With a
who wants to produce a flying model which is quite different rubber-powered model a simple crank mechanism can be used,
from all other types. connected by a link rod to pivoted outer panels of the wing
(Fig. 50). A more or less conventional tail unit is added for

Actually the autogiro kite is a simpler type to make, and

usually more successful, than a powered autogiro (Fig. 49). The
powered model can use a similar configuration, but the con¬ stability, the fuselage normally being of elementary form to
struction should be lighter—e.g. using a built-up box fuselage accommodate the rubber motor and carry the tail. Construc¬
and built-up tissue covered rotors. Whatever form of power is tion must be kept as light as possible and conventional rather
used (rubber motor or engine), this should be mounted with than bird-wing shapes are most likely to prove successful—at
a “down-thrust” angle of some 10 to 15 degrees. least for the first models when design details are being worked
Another quite tricky model is the ornithopter or “wing out by “trial and error.”
flapper.” In this case propulsion is obtained by a flapping wing A far more promising layout in the unorthodox field is the
in a similar manner to bird flight. What works very effectively flex wing—virtually a type of powered kite which has been
for birds, however, is extremely difficult to emulate in model produced as full-size flying machines. A basic model flex wing
flight and the best of ornithopter models only has “marginal” consists of nothing more than three stiff ribs or spars (e.g.
power of flight. The majority never have enough power to aluminium tube or bamboo) assembled in arrowhead form and
accomplish more than a prolonged glide. Again, however, it is a loosely covered with a light, non-porous flexible material such
 AEROMODELLING UNORTHODOX MODELS 83
82

as sheet polythene. The material should be slack enough to direction to normal simply by winding the motor up the other
“balloon” evenly between the spars (Fig. 51). This can be way.
most easily done by attaching the covering taut to the arrow¬ The canard or tail-first model also makes an interesting pro¬
head frame and then bending the outer spars inwards an equal ject, and is usually employed with a pusher propeller (although
amount each, as shown in the diagram. canards also make interesting gliders). Typical design pro¬
The model is then completed by suspending a conventional portions are shown in Fig. 53. It is a characteristic of all
fuselage below the wing. This suspension should be rigid (e.g. canards that the balance point comes in front of the main wings,
wire braced) so that the wing is mounted at an angle of about
15 degrees to the fuselage and lined up accurately fore and aft

Fig. 52

with the wing centre spar. Correct balance point will then come
about 40 per cent back from the front of the “arrowhead” with
an effective wing sweep of about 50 degrees (farther back if the
sweep angle is greater). The flex wing model can be powered
by a rubber motor or a small engine, or can even be flown as
a glider.
Many of the other unorthodox models are almost conven¬
tional by comparison. Thus the pusher, for example, is a more or with the leading plane always set at a greater rigging angle
less conventional flying model layout with the airscrew at the (angle of incidence) than the main wings. Properly propor¬
rear of the fuselage instead of the front (Fig. 52). Proportions tioned, the canard arrangement is quite stable and efficient
are orthodox free-flight, except that the wing must be located and can even make a good “duration” design for rubber-
farther aft or the fuselage nose lengthened to balance the powered models.
weight of the rear propeller (and engine in the case of a power Tail-less models have always been popular, although they
model). The undercarriage will also have to be relocated to present considerable stability problems. For greatest efficiency
protect the propeller in landing. Also, of course, the propeller the wing needs to be of fairly high aspect ratio with a good span,
must be a “pusher” type, carved with opposite pitch or when stability is provided by sweepback and reflexing or de¬
rotating in the opposite direction to convention. A “pusher” creasing the angle of incidence towards the tips. Efficiency
propeller is required on engines which normally rotate only suffers because reflexing reduces the overall lift, also rather
in one. direction (anti-clockwise when viewed from the front more stable aerofoils (and thus less efficient lifting sections)
or propeller end). A conventional propeller can be used on a normally have to be employed. Stability provided by reflexing
rubber-powered model since this can be driven in the opposite and sweepback is also far less than that given by an orthodox
84 AEROMODELLING
UNORTHODOX MODELS
85
tailplane. Thus “duration” performance cannot compare with
leading edge. The engine is most conveniently carried on a
an orthodox layout.
short fuselage or nacelle extending from the front of the cir¬
The high-aspect-ratio tail-less layout is the logical choice for
cular wing, with directional stability provided by a fin of
a glider or rubber-powered model. With engine power (and
suitable size near the back of the wing.
more especially “Jetex”) the wing planform can be rendered
Saucers, deltas, and to a less extent canards, also make
as a true delta. This will considerably improve the stability,
suitable control-line models, the tail-less designs in particular
but still further reduce the efficiency or amount of lift generated
being highly aerobatic with a hinged elevator at the wing
trailing edge. The autogiro is another possible control-line
type.

FLAT BOTTOM
WING SECTION

Although not, perhaps, “unorthodox” models in the same

sense as the models previously described in this chapter,
electric-powered models do come into this category because of
by a given wing area. Torque becomes a problem if too their comparative novelty. Electric motors are not normally
powerful an engine is fitted, so the delta is really best suited to suitable for model aircraft power units because of their low
a low-powered sports model. Increasing the span of the delta— power/weight ratio, and also, because they have to carry
i.e. using a broader triangular planform—is no answer since additional weight in the form of batteries. However, small
whilst this may improve lifting efficiency it will also decrease lightweight motors powered by very light batteries have been
stability.
used successfully for free-flight models. The chief requirements
An alternative method of approach is the purely circular are that the model must be extremely lightly constructed
wing planform or flying saucer (Fig. 55). Strangely enough this (usually on rubber “duration” lines), with the motor driving
makes an excellent free-flight model with small engine power a large-diameter propeller via a high reduction gear ratio.
(although not suited to rubber power), with relatively few This gives a better performance than the same motor driving a
stability problems. More or less conventional wing sections can small-diameter fast-revving propeller direct and also reduces
be used, but not undercambered aerofoils, with a balance point battery consumption.
between one-third and one-half the chord back from the Merely adapting a conventional rubber model to electric
86 AEROMODELLING

motor power will not produce successful results. The design

requirements are highly specialized and the power available
will, in any case, be marginal. The electric-powered free-flight
model is, therefore, essentially restricted to still-air flying.
chapter 9
With an efficient low-consumption motor battery power can
be provided by pen cells (preferably the high-energy type) with
CONSTRUCTION
a motor running time of up to io minutes or more from a single
set of batteries. It usually becomes necessary to limit the motor
Built-up construction is favoured for the majority of small to
run by a cut-off switch operated like a dethermalizer (see
medium-size models, or for all components on larger models
Chapter n). With smaller models the new ultra-lightweight
where it is necessary or desirable to build down to minimum
water-energized batteries may be employed. These weigh
weights. Regardless of the component—wings, tail, or fuselage
something less than i/ioth ounce each and are capable of
—the basic framework is always assembled flat over a full-size
giving i .4 volts per cell, once activated, for a matter of up to
drawing or plan, spars, etc., being pinned in position over the
one minute. After that the cell is exhausted and must be
plan with other parts cemented in place—again using pins to
thrown away.
hold in position if necessary. The framework is then left in
Successful designs for unorthodox models are almost always
position until the cement has quite set.
evolved on a “cut and try” basis, often through a whole series
The building board used must be flat and true for accurate
of similar models, some successful and others not. The best
construction. The plan is laid over the board and, to protect
starting point is to build from, or base an original design on, a
the surface, can be covered by a sheet of waxed paper or a
plan of a successful model of the same type. Lacking mass
candle rubbed over the plan surface to prevent cement sticking
appeal, few unorthodox models are produced as kits, but plans
to it. Assembly of different components then follows a definite
are readily available of almost any type, many of which may
pattern and can be best dealt with under separate headings.
be record-breaking of record-holding models in their class.
The chief sources of such plans in this country are—
Aeromodeller Plans Service, 38 Clarendon Road, Watford. Fuselage Construction
Model Aircraft Plans, 19-20 Noel Street, London, W.i.
In the case of a simple built-up box fuselage the two sides are
built flat as separate frames and subsequently joined with
formers and/or spacers. The basic stages involved are shown
in Fig. 56, which can be studied in conjunction with the
following notes.
The four main spar members or longerons should be selected
to be as near equivalent as possible in grade. This is important
to minimize the risk of the fuselage “springing” out of shape
when finally assembled. To ensure that both side frames are
identical it is also best to build them one on top of the other
directly on the plan. This is also the quickest method.
The longerons are first pinned out in position accurately over
the plan. Spacers are then cut accurately to length, in pairs,
87

i
88 AEROMODELLING
CONST RUCTION 89

and cemented in place. Cement in place one station at a time maining spacers can be cut and fitted to complete the basic
rather than cut two complete sets of spacers and then cement framework. Again cut the spacers in pairs (top and bottom)
in place. Insertion of the spacers may “spring” the outline and work one station at a time. Check the fuselage assembly
slightly, so by measuring off spacer lengths one station at a for truth and squareness and allow to set.
time, and fitting, you will get the most accurate assembly. It then only remains to add the detail parts, which may be
After coe, dieting the frames they should be left pinned down specified, such as a nose former facing, anchor e plates for
for several hours to ensure that the cement has completely set. dowels, etc., bind the undercarriage in position (where appli¬
cable) and so on. On some models “fill-in” sheeting may be
called for, which has to be cut accurately to size and cemented

in place. Complete all the fuselage details at this stage and then
clean off and sand down lightly ready for covering.
Then remove all pins and lift the frames off the plan. They will
There are, of course, variations on this type of construction,
be stuck together, so separate carefully by running a knife
but the same basic technique usually applies. For example,
between them, taking care not to cut into the wood.
formers cut from sheet balsa or even ply may largely replace
Joining of the two sides should always start at the middle or
spacers. This will, in fact, make it easier to get the assembly
widest section of the fuselage. Cut the spacers to required
square, although it usually increases the weight of the fuselage.
length (measured off the top view of the fuselage), or use
In some cases, where the fuselage section is rounded, a com¬
formers, as appropriate. Sides and spacers can then be cemen¬ plete half shell may be built flat over the plan, comprising
ted and pinned together if the longeron size is J in. square or
backbone or keel members to which are cemented formers,
larger; otherwise hold the assembly with two rubber bands. with stringers then added—Fig. 57. This assembly is removed
Block up square (as regards cross section) and true (as regards
when set and the other half of the shell built directly on to it.
alignment over the plan view) and leave to set. It is very In other cases a basic frame known as a crutch may be as¬
important to get the fuselage assembly true at this stage. sembled over the plan view rather than the side view, with
The next stage consists of joining the tail ends of the sides formers and stringers then added to complete the whole of the
(usually these are cemented together); and the front (with top of the fuselage whilst the crutch frame is still pinned down
spacers). Use pins again to hold in position, and once more
to the plan. The bottom fuselage construction is then added
align over the plan view. When the cement has set, the re- after the main assembly is removed from the plan.

-
9° AEROMODELLING CONSTRUCTION
9i
Wing Construction tip is at the required tip-rise measurement, this being twice the
Simple wing construction is shown step-by-step in Fig. 58. tip-rise value for dihedral expressed at one wing tip.
Either the complete wing is built as one flat panel in the case The same principle applies when forming a polyhedral joint
of small models, or each half wing built as a flat panel. Leading in a wing panel. The main thing is to keep one panel (or part
and trailing edges and the mainspar(s) are pinned down panel) pinned down and block the other up accurately whilst
directly over the plan and the ribs then cut and cemented in the joint is completed and allowed to set. Blocks used to prop
place. If the wing tips are also built up, these are also cemented up the second panel must be at right angles to the leading and
in place at this stage. Alternatively, the wing tips may be
carved from light block balsa, added after completing the main

trailing edge (or mainspar). If slewed they will cant the panel
slightly and alter its angle of incidence.
Certain complications arise in the case of “duration” model
wings where the section is usually undercambered. To make
sure that the spars line up with the aerofoil section it may be
necessary to block up both the mainspar and the front of the
trailing edge when pinning these members down initially over
the plan (Fig. 59). Similar considerations also apply when
building a symmetrical-section wing (such as used on control¬
line models and some radio-control models). Here it becomes

Fig. 58

wing assembly. Spars which slot into the top of the ribs are Fig. 60
added after the ribs are cemented in place.
To form the dihedral a one-piece wing panel is notched at necessary to block up both the leading and trailing edge
the dihedral joint, one panel supported so that the tip has the (Fig. 60). If a single spar is used running through the centre of
required amount of dihedral (tip rise) and dihedral braces the ribs, the ribs must be assembled on to this spar and the
added to strengthen the joint. In the case of two-panel wings whole lot then positioned over the leading and trailing edge
the spar ends are carefully angled for a neat fit and the wing pinned down to the plan. Sometimes this is avoided by cutting
panels joined at the correct angle with dihedral braces and well- symmetrical ribs in two parts so that the mainspar can be
cemented joints. Gussets are also usually added to strengthen pinned down on to the plan for accurate alignment, the
the joint at the leading and trailing edges. Once again one wing “missing” part of the rib being added later after the main
panel is left pinned down and the other blocked up so that the assembly has been completed and removed from the plan.
92 AEROMODELLING CONSTRUCTION 93

Accurate alignment of ribs is helped by notching the leading board. This will eliminate the risk of warps being accidentally
and trailing edges so that the rib ends actually locate in these worked into the wing whilst applying the sheeting. Wing
notches. This calls for an extra length of rib—thus notched panels are joined in the same manner as before, but extra care
assembly cannot be applied to a kit model which does not is needed to shape the ends of the sheet accurately for the
provide for it and where the ribs are already die-cut. Also neatest possible joint line at dihedral breaks.
mainspars should never be notched to assist alignment as this Some wing designs may employ additional diagonal bracing
to provide an “anti-warp” structure—see Fig. 62. Basically, this
additional bracing simply increases the rigidity of the wing as
regards twisting loads, such as those imposed by the tautening
of the covering. It is imperative that any such diagonal bracing
Fig. 6 i be fitted before the wing panel has been removed from the
building board. If added later it will almost certainly push the
will weaken the spars unduly. The weakening effect on leading wing out of shape and add a permanent “built-in” warp.
and trailing edges is negligible; but where notches are employed Other wing designs may incorporate anti-warp features in the
they should always be cut with a flat file of the same thickness arrangement of the ribs, e.g. “W”-alignment instead of straight
as the rib, not with a knife blade. fore and aft, or crossing in “X” form (known as geodetic).
On larger models a slightly different form of construction
may apply. To save weight, the trailing edge may be built up
of two pieces of sheet balsa, rather than a triangular section of Tailplane Construction
solid balsa. Also the leading edge of the wing may be sheet
Assembly of a built-up tailplane is basically the same as that
covered (Fig. 61). Where this applies, as much sheet covering
of a wing, with the complete frame built as one flat panel.
as possible should be fitted and cemented in place with the
Tailplanes are not usually given any dihedral angle and so the
basic wing framework still pinned down flat on to the building
question of cutting and making a dihedral joint does not arise.

WARREN GIRDER GEODETIC

Fig. 62 Fig. 63
94 AEROMODELLING CONSTRUCTION

Spar and rib arrangement is usually similar to that of the wing,

so exactly the same considerations apply. Tailplane spar sizes PACKING

will be smaller and lighter, however, so particular care may be PACKING

needed to ensure a warp-free structure. Symmetrical tailplane
sections set a problem in this respect and are generally best
built with a “split” rib section (Fig. 63).

Shaping Leading and Trailing Edges

Leading and trailing edges are normally shaped from solid

balsa spars. These spars may be pre-shaped—e.g. supplied or
bought as leading and trailing edge sections, respectively—or must be made to avoid warps since this can lead to a lot of
individually shaped from square or rectangular strip balsa. trimming troubles on the completed model. A flat-section fin is
In the latter case it is always best to shape these sections before the simplest to make, but the one most prone to warping when
pinning in place over the building board rather than carving tisue covered, especially if covered on one side only as is often
and sanding to a matching shape after the frame has been recommended with small models. It is always best, in fact, to
completed. This is because shaping, especially on the trailing cover any fin on both sides, and strictly necessary if the section
edge, will tend to warp the strip. Thus if a wing is built with a is “streamlined.”
rectangular section trailing edge which is subsequently carved A fairly straightforward method of getting accurate assembly
and sanded down to a matching triangular shape after com¬ in the latter case is to block up leading and trailing edges
pleting the wing or tailplane framework it will have a natural equally, as shown in Fig. 64. The ribs can then be rectangular
tendency to develop an upward curl. If the structure is not strips which just rest on the plan surface, making for good
rigid enough to withstand this, it will simply mean that the accuracy of alignment; or they can be cut to symmetrical
wing frame will warp under the locked in stresses of the trailing section ready for assembly. In the former case the symmetrical
edge tending to make this member bow upwards. This may not section is formed by sanding away the ribs after the frame has
be very significant or even important on a sports model or a been built.
fairly ruggedly built wing, but it can be important on a light¬ Fins cut from light sheet balsa are quite suitable for glider
weight duration-model wing. and power models (although rather too heavy for lightweight
A certain amount of final shaping will probably be neces¬ rubber models). The fact that they are solid does not mean
sary in any case after the wing frame has been removed from
the building board. If the trailing edge has to be sanded to any CEMENT JOINT

extent, and the structure itself is a lightweight one, this can be

compensated by sanding the underside of the section lightly
to pull out the “bow” which may be induced.
HARD BALSA INSET

Fins SOFT QUARTER GRAIN

SHEET BALSA

The fin is basically a simple structure which should not

present any particular difficulties, although every endeavour
96 AEROMODE L LING TABLE XV. SUMMARY OF MATERIAL SIZES, etc.
FOR ALL TYPES OF MODELS
that they will be warp-free, however. In fact, it is generally
better to inset a strip of harder grade balsa with the grain at
right angles to the main sheet to provide anti-warp properties—
see Fig. 65. This method of inserting a “key” piece is also
particularly useful as reinforcement when the fin is of such a
size that it has to be cut from two pieces of sheet cemented
together, the key then being equally disposed on either side
of the joint line.
With commercial model aircraft designs—kits or plans—
building instructions are usually specific, especially for parts
which differ from conventional practice. Where no such in¬
structions are given, assembly will normally follow on the lines
detailed in this chapter.

7
AEROMODELLING CONSTRUCTION
99
IOO AEROMODELLING CONSTRUCTION

FUSELAGE TZHT
NOTES:
N.R-NOT RECOMMENDED
N. S' NOT SUITABLE
H-HARDWOOD
P-PLY
 COVERING AND FINISHING 103

CHAPTER IO

COVERING AND FINISHING

The starting-point for any good covering job is a “clean”

framework. That is, all joints, etc., in balsa frames should be
smoothly blended, spacers flush with longerons, no blobs of
surplus cement standing up clear of the outline, and so on. All
balsa surfaces, etc., should also be smoothed by sanding with
“flour” glasspaper. Sanding should never be attempted too
vigorously on balsa frames, however, for it is all too easy to
sand away too much wood, cut notches in spars or even break
relatively fragile components like ribs.
Tissue covering may be stuck to the frames with tissue paste
or tissue cement, or with dope. In the former case the paste or
tissue cement is applied directly to the framework (preferably
with a short stiff-bristled brush) and the tissue panels laid in flush with a razor blade. Only a new razor blade will be sharp
place. Using dope, the whole frame is given a coat of thick enough to make a neat job of trimming, and it should be
dope and allowed to dry. The tissue is then laid in place and rinsed occasionally in water to remove paste which will tend
dope-thinners are brushed through the tissue lying over the to adhere to it.
doped frames. This is a much more tedious method of covering Having completed covering of one side, the three remaining
and requires patience and no little skill to produce consistent panels can then be covered in turn, in exactly the same manner.
adhesion. It does, however, produce a neater job. Paste (or The main aim is to get the tissue applied smoothly and evenly
tissue cement) adhesive is normally recommended. without wrinkles, and firmly stuck down around the edges.
Stages in covering a typical box fuselage are shown in Covering should not be stuck to intervening spacers, etc., only
Fig. 66. Four tissue panels are required, one for each of the the outline. Also it is not necessary to pull the tissue covering
sides, top and bottom, each cut at least i in. oversize all round. absolutely taut. It is far more important to get the covering
One surface is then covered at a time. Start by pasting the applied without wrinkles.
longerons at the middle of the fuselage. Then lay the tissue Covering wings is a little more tricky, but exactly the same
panel in place, pull reasonably taut and smooth down on to principles apply. Separate pieces of tissue are required for the
the pasted longerons. Proceed to paste a further length of top and bottom surface of each wing panel (it is impossible to
longeron, top and bottom, for a further two or three bays, cover neatly with one piece wrapped around the leading edge),
pulling and smoothing the tissue in place. Work first to the and a separate piece is needed for each panel bounded by a
front; then start from the middle again and work to the rear. dihedral break (it is impossible to carry covering over a
Surplus tissue overlapping the edges can then be trimmed off dihedral joint).
102
 104 AEROMODELLING

Paste is applied to the leading and trailing edges only, start¬

ing at the centre of the panel. Then work towards the tip and
centre, 3 or 4 in. at a time, just as in covering a fuselage side.
Alternatively, attach the tissue first to one end of the panel
across a rib, then pull taut and stick down to the other end.
Now work from the centre in each direction, sticking the tissue
to the leading and trailing edge. Again be sure to pull out all
wrinkles and aim for even tension rather than overall tautness.
Trim off surplus tissue at the edges with a razor blade before
Fig. 68

for such a job. Tissue cement or thick dope must be used.

Covering procedure is then to attach the covering to two or
three ribs in the centre of the wing panel first, then to work
along the length of the wing sticking to each rib in turn and
pulling out taut (Fig. 68). Having achieved this, apply paste
to the leading and trailing edges and tips and complete
securing the covering to the outlines.
COVER BOTTOM FIRST
Tailplanes are covered in exactly the same manner as
wings, except that dihedral breaks and undercambered ribs
are seldom encountered (Fig. 69). Hence one panel of tissue
can be used for covering the bottom surface and another panel
for covering the top surface—with special treatment for the
starting to cover the other side of the panel. Bottom wing panels, top tips, if necessary. Covering built-up fins is simpler still,
being flat, are easiest to cover. Upper surfaces, where the tissue using a single piece of tissue for each side and pasting down to
has to be pulled out evenly over curved ribs, are a little more the outline only.
difficult; but always cover each bottom panel with the appro¬ Tissue-covering can then be tautened by spraying or painting
priate top panel rather than dealing with the bottom panels
first and then with the top panels.
Covering the tips may present a special problem since it may
be impossible to pull out wrinkles in the top panel. In that
case, terminate the top covering by pasting it on to the last
rib and cut a separate piece of tissue to cover the tip. Working
with these smaller pieces you should be able to produce a
wrinkle-free covering. In all cases covering is pasted down
to panel edges only, not to individual ribs and spars as well.
Wings with undercambered ribs need special treatment for
covering the bottom surfaces. Here the covering must be
attached to each individual rib and tissue paste is not suitable Fig. 69

*v*‘**t
/
io6 AEROMODELLING
COVERING AND FINISHING 107
with water. Spraying is best since a brush can readily tear wet
dopes should not be used, except on the larger and heavier
tissue (especially Japanese tissue or lightweight tissue). If a
sports-type models where weight is not so important. It is
brush has to be used, choose a very soft mop brush and apply
better to provide colour by using coloured tissue, even on
very light strokes only.
flying-scale models.
On drying out the tissue will tauten appreciably, resulting in
Since clear dopes tauten on drying, the same precautions
a smooth, tight covering job if (i) the tissue panels have not
must be observed as when water-spraying tissue. As soon as the
been applied too slackly in the first place; and (ii) the covering
dope has dried to the extent that the surface is no longer
has been applied without wrinkles. Wrinkles are the usual fault,
tacky, pin or weight down wings and tailplanes on a flat
for bad wrinkles left in the initial covering, as applied, will
surface and leave at least overnight. If this is not done, bad
never pull out; smaller wrinkles may. A good appearance to
warps may be produced as the dope dries. It is possible to
the finished, tautened covering, therefore, depends on “wrinkle-
remove these warps by gently heating the surface affected (e.g.
free” application in the first place.
holding in front of a fire) and then twisting it true and holding
Tautening may also introduce another problem—warping
in this position until the covering has cooled. There is always
of the framework. This applies only to wings and tail units,
the distinct likelihood, however, that warps removed in this
never to fuselages. After wings and tailplanes have dried off to
way will show up again later, and it is better if warps can be
a “damp dry” state, therefore, they should be laid on a flat
avoided entirely by taking the necessary precautions at the
surface and weighted down so that during final tautening they
covering and finishing stages.
cannot be pulled out of shape. This is particularly important
With silk or nylon covering, the basic technique is somewhat
in the case of lightweight structures—is essential, in fact, if
similar, although the material may be rather more difficult
warping is to be eliminated.
to get to stick down smoothly. Such materials are most easily
Tautened covering adds appreciably to the strength and
applied in a damp state—i.e. the panel of covering is soaked
rigidity of the framework. The covering itself, however, is
in water, wrung out, and then applied whilst still damp. It is
relatively weak and is not suitable for normal handling. It
also necessary to apply such materials quite taut since they
therefore requires doping to provide additional strength and a
cannot be water-shrunk after the covering is completed, and
reasonable degree of moisture resistance. The standard material
final tautening must be done with dope. Thus considerably
for this is clear model dope, which itself will normally have
more skill is involved in pulling a silk or nylon panel in place
certain tautening powers; thus any slackening of the covering
tautly over the frame, and without any wrinkles appearing.
under the wetting action of the dope is more than taken up
It may also be expedient, or necessary, to use pins to help hold
again by its own tautening powers.
the material in place as covering proceeds.
To avoid excessive tautening, with a further danger of
Nylon (or silk) covering must be doped initially with full-
warping as well as tending to make tissue-covering brittle, clear
strength clear dope. Ordinary model dope will not have enough
dope should be mixed with an equal proportion of thinners.
tautening power and may even introduce slackness in what
At least two coats should be applied on wings and tailplanes
was reasonably tight original covering. Apply one, two, or
allowing time to dry between each, and up to four coats on
three coats of full-strength dope, as necessary, to produce a
fuselages. More coats may be necessary to fill porous open-
good taut covering. Any further dope coatings to seal can then
weave tissues or on larger models employing a heavier tissue
be thinned down, if necessary, to save adding too much weight.
covering. Several thin coats are better than one or two coats of
Again coloured covering is best introduced by using coloured
thick dope, but too many coats of dope will only add unneces¬
material, although this can be improved on by using a mixture
sary weight and make the tissue covering more brittle. Coloured
of clear and thin colour dope (the same colour as the material)

-
io8 AEROMODELLING

for final coats. Solid colours should be restricted to trim,

unless model weight is not all that important.
Conventional model dopes are reasonably resistant to diesel
fuels, but are attacked and softened by glow-motor fuels.
Power models, particularly those with glow engines, therefore
require an additional coating of “fuel proofer.” This is basi¬
cally a clear varnish-type finish, resistant to fuels, which is

: l.w.-lightweight; m.w. = medium-weight; h.w. = heavy weight; w.s.=wet strengthened.

applied as a final finish over a normal doping scheme. One
overall coat of fuel-proofer is normally all that is required in
this respect. Butyrate dopes, on the other hand, are fuel-
resistant and can be used throughout without final fuel¬
proofing. The two types should not be mixed, however, so use
either normal dopes throughout (followed by a coat of fuel-
proofer) or butyrate dopes throughout.
The covering of sheet balsa surfaces requires a slightly dif¬
ferent technique from that previously described for frames. The
covering material (tissue, silk, or nylon) should be stuck down
all over the sheet and not just to the edges. If not, the covering
will invariably wrinkle. Dope is the best adhesive here for
tissue covering, and gives the neatest job at the least weight,
but either dope or paste can be used equally well with silk or
nylon covering over sheet balsa. Finishing then follows similar
lines to that already described, except that there is no point in
trying to water-tighten all-over tissue covering applied to sheet
balsa unless there are some “dry spots” showing because of
lack of adhesion in certain areas.
In the case of control-line models, particularly sports type
and speed models, weight is not particularly important and
more attention can be devoted to getting a first-class finish,
using coloured dopes. With solid balsa surfaces the most im¬
portant part is complete filling of the wood grain, using re¬
peated coats of dope or special grain fillers and sanding down
perfectly smooth between each coat, using nothing coarser
than 300 grade “wet or dry” paper for the final stages. Similar
treatment may also be advisable on sheet balsa surfaces on
larger free-flight models which are to be tissue- or nylon-
covered, to reduce any grain roughness.
Final finishing of control-line models then follows a
similar procedure to that used for automobile work. Having

I
I IO AEROMODELLING

satisfactorily filled the grain and obtained a glass-smooth, non-

porous surface on which to work, surface imperfections re¬
maining are filled with stopper and rubbed flat, followed by an
undercoat or base coats, as necessary, again flatting between
each coat. Finishing coats are then applied, with or without
intermediate flatting, and perhaps finally buffed and polished
after allowing ample time for the finish to dry hard. Applica¬
TRIMMING AND FLYING
tion of the various coats by spray is more or less essential in
order to obtain a true “professional” finish.
The rigging of a model for flight is largely a matter of ensuring
Spray application is also to be preferred for the doping of that the wings and tailplane are set at their correct incidences
any type of model since it produces a smoother, more consistent
or rigging angles, with the balance point of the model at a
appearance than brush painting, even with clear dopes. How¬
suitable place. Also involved in rigging is accurate alignment of
ever, brush painting is usually considered quite satisfactory for
the bulk of free-flight models.

CHECK FOR WARPS

AND MISALIGNMENT

SIGHT MODEL FROM REAR

Fig. 70

the wings and tail relative to the fuselage, the fin aligned true
and square, all surfaces being free from warps, etc. These are
largely constructional features which need checking out before
attempting a first flight with a new model.
Accuracy of alignment is usually done by eye, judging
whether or not the assembled model looks true. In particular,
warps will readily show up if the model is sighted from the
rear, looking directly towards the trailing edge of the wing or
tailplane (Fig. 70). Warps, or any tendency for the tailplane
or fin to be out of square will usually show up clearly. These
are basic rigging faults which need correcting.
To check whether the wings and tail are square to the fuse-
lage, and parallel to each other, the simple measuring tech¬
nique shown in Fig. 71 can be used. If accurately aligned,
measurement Ai should equal A2; and measurement B1
should equal B2.
The balance point of the model is almost invariably referred
 AEROMODELLING TRIMMING AND FLYING
112 ”3
should always be conducted in calm weather, preferably dead
calm with no wind at all. If there is a light wind, then launch¬
ing must always be done dead into the wind. It is also useless
to attempt trimming in “calm” areas on windy days, such as
in the lee of a building, etc. The air in such a region is far from
calm, and serious damage to the model is likely to result from
the upsetting effect of turbulent air causing it to crash.
Gliders are the simplest type of free-flight model to trim

Fig. 71

to its distance back from the leading edge of the wing, that is,
the model should balance level if supported under the wings at
this point. The balance point, or centre of gravity as it is some¬
times called, may or may not be marked on the plan. Its Fig. 73

position is dependent on the layout of the model and also on

since they have to be balanced for only one set of conditions—
gliding flight where the “power” is provided by gravity and the
model moves forwards and downwards at a gliding angle
determined by its aerodynamic layout and its trim. Glide
angle is not affected by weight of model, although the heavier
the model the faster it will fly, so its time of descent from a given
BALANCE LONGITUDINAL DIHEDRAL =-A°~B°
POINT
height will be shorter.
Hand-launching can be used to establish a rough initial
Fig. 72
trim, launching the model from shoulder height and aiming at
the longitudinal dihedral between the wing and tailplane a point on the ground about 20 to 30 ft. in front. The knack of
incidences (Fig. 72). The latter is difficult to judge or measure, successful hand-launching is to release the model at the correct
but is usually fixed by the shape of the fuselage and wing flying attitude and speed—not nose-up or nose-down, and not
mount (where applicable). The smaller the longitudinal dihe¬ too fast or too slow. Hand-launching is only good for rough
dral (i.e. the smaller the difference between the rigging angles trimming, but it can show whether a model is badly out of trim
of the wings and tailplane), the farther aft the corresponding or not; then necessary adjustments can be made before pro¬
balance point. It is not vital to know exactly where the balance ceeding to high-start launches.
point is, or the exact value of the longitudinal dihedral, since Assuming that a glider has stable proportions to start with,
both may be adjusted in trimming to arrive at a suitable balance its initial trim, as rigged, may give it a stable glide, a stalling
of forces for stable flight. . flight, or diving flight (Fig. 73). If it stalls, then this over¬
The first flights of any new model—and all trimming flights— elevated condition may be corrected by altering the trim by
8
 TRIMMING AND FLYING
II4 AEROMODELLING ”5
ticular trim. The angle of glide may be quite flat, or at the
either (i) adding more weight to the nose (equivalent to moving
other extreme so steep that the model is really diving rather
the balance point farther forward), (ii) shifting the wing posi¬
than gliding. In other words, there are degrees of trim, corre¬
tion backwards (which has exacdy the same effect as (i); (iii)
sponding to different glide angles. If we start with stalling
reducing the rigging angle of the wing; or (iv) increasing the
flight and add a little nose-weight at a time the glide will pro¬
rigging angle of the tailplane. Both (iii) and (iv) have the
gressively become flatter until eventually a point is reached
when further weight will start to steepen the glide again. Add
TABLE XVII. more weight still and the model will be diving. All the time
TYPICAL RIGGING DETAILS FOR FREE-FLIGHT MODELS
with these trim changes the flying speed of the model will be
increasing.
Wing Balance Wing Tail
Model Type Position Point Incidence Incidence
(% Chord) (Degrees) (Degrees)

Gliders
Cabin High wing 35-40 3 0
Contest High wing 50-66 3 to 2 0 to 4-4
Rubber r High wing 35-40 3 to 3i 0 to — 1
Sports
L Shoulder wing 35-40 34 O to —I
Rubber r Shoulder wing 35-40 34 0
Duration 1 High wing 40-50 3 0
1 '
Pylon wing 60-75 3 +1
Power High wing 35-40 34 0 Conversely, if we started with a model which was under¬
Sports ] Shoulder wing 30-35 3 0 to —I elevated and diving rather than gliding, reducing the nose
Low wing 30-35 3 0
weight to move the balance point farther aft, or increasing the
Power
longitudinal dihedral angle, would correct this condition. The
Duration Pylon 66-75 2 to 24 4-4 to +1
glide would progressively become flatter, then start to steepen
again, and shortly afterwards, with further over-elevating trim
Note: Having established the balance of the model at the nominal or design
position, trimming is most usually carried out by adjusting the tailplane incidence
adjustment, the model would start to stall. This time, with
by packing under the leading or trailing edge. small changes in trim, the flying speed of the model will be
decreasing each time.
effect of reducing the longitudinal dihedral angle, which is the The two limits of trim which are of interest are the trim for
same thing as applying an under-elevating trim effect to counter flattest glide and the trim for minimum flying speed (Fig. 74). With
a balance point which was too far aft to start with. However, the former trim the model will fly farthest from a given height.
reducing the longitudinal dihedral angle reduces the stability, With the trim for minimum flying speed, without actually
so methods (i) or (ii) are safer trim techniques. If the wing stalling, the model will have a steeper angle of glide but it
position is fixed by the fuselage shape, only (i) can apply; this will descend more slowly because it is flying at minimum
is the usual method of trimming gliders. speed. Thus it will give the longest duration of flight launched
Having achieved a stable glide trim another interesting fact from a given height. It is thus called the trim for minimum
sinking speed.
now emerges. Gliding flight does not correspond to one par-
Il6 AEROMODELLING
TRIMMING AND FLYING II7
In practice, this trim is quite easy to find. The trim is ad¬
justed, a little at a time, until the model just begins to stall. A does not give enough height for a reasonable glide, try half
little extra nose-weight is then added to bring the trim just turns, or even more. If the model shows the slightest signs of
below the stall, and this will correspond very closely to the stalling under power, add a piece of packing between the top
minimum sinking speed for the glide. This, in fact, is the basis of the noseblock and the front former to give some downthrust
of “duration” trim on all free-flight models as far as the glide (Fig- 75)-
is concerned. It does not necessarily apply to sports-type Work on the glide trim through a series of half-turns flights
models, however, where a flatter glide, corresponding more until satisfied. The power trim can now be dealt with. Basically,
nearly to the trim for flattest glide may be preferred. This is this involves packing the noseblock downwards slightly (down-
because the slightly greater degree of under-elevation may be thrust) to correct any stalling tendency under power; and to
beneficial to prevent the model actually stalling in gusty air.
Glide trim can only be observed correctly when the model is
gliding from an appreciable height, e.g. after tow-launching in
the case of a glider. Adjustments are then made, a little at a
time, and re-checked by a further “high-start” flight. It is
impossible to adjust the glide trim of any model properly by
hand-launching near the ground. Change in directional trim
will also affect the glide trim. Thus any adjustment to make a
model turn more sharply will produce a corresponding under-
elevating effect; and conversely straightening out a turn will
produce an over-elevating effect. One method of curing a Fig. 75

tendency to stall on the glide, in fact, is to make the model

describe a slightly tighter turn. This must never be overdone, the right (sidethrust) to counteract torque and make the model
however, since too tight a turn will readily finish up as a spiral circle to the right on the climb (this being the safest and most
dive into the ground. The under-elevating effect of a sharp efficient power-on trim with a propeller revolving in the con¬
turn is as powerful as that! ventional direction).
In the case of power models, the correct trim technique is Neither amount of packing should be overdone, and adding
first to establish the required glide trim and then to adjust the side-thrust increases the effectiveness of down-thrust. If over¬
power trim. At the same time, however, the overall trim has to done, the model will go into a spiral dive to the right instead of
be fairly near the mark to start with to get the model to fly at climbing. Side-thrust should be limited to a maximum of 3
all. Thus initial hand launched glides can be attempted to degrees (equivalent, usually, to not more than 1/32 in. packing
establish a reasonably “safe” glide trim, preferably carried out on a small model or 1/16 in. on the largest size of rubber model).
over long grass to cushion the model from heavy landings More, or less, down-thrust may be required. It is safest to use
should it be badly out of trim to start with. The next step is little or no side-thrust and do all the final trimming on down-
then to get the model up to a sufficient height under power for thrust ; but most efficient to use the amount of side-thrust which
the true glide to be observed and final glide trimming attemp¬ gives the required right-hand circle on the climb allied to the
ted, as required. minimum amount of down-thrust to prevent any tendency to
stall.
In the case of a rubber model this is fairly straightforward.
The motor is wound up to about one-third turns only. If this In making side-thrust and down-thrust adjustments, the
number of turns applied to the motor is progressively increased
 TRIMMING AND FLYING ”9
118 AEROMODELLING
a spiral dive under power, particularly if this turn is given by
each flight. Trim is not complete until the model has been
offsetting the fin or a rudder tab. Turn adjustments of this
flown on full turns and the corresponding side-thrust/down-
type should be reduced to an absolute minimum. In fact, it is
thrust adjustment established. This may also change if the
usually best to produce turn trim by some other means, es¬
motor size is altered, but otherwise can be regarded as per¬
pecially on high-powered models. Devices used include trailing
manent. All packings used for trim can then be cemented
“drag” flaps on one wing, automatic timers for kicking a rudder
permanently in place—this applying also to any packing used
over for glide turn after holding it straight during the power
under the tailplane (or wings) in establishing the glide trim.
flight, and tailplane “tilt.”
The latter is quite effective and consists of packing the tail-
plane up so that it is tilted relative to the wings (Fig. 76). This
will induce a tendency to turn the model in the direction of the
higher tailplane tip and is far less drastic in action than any
rudder tab or similar control. Its use is, however, largely con¬
fined to the trimming of contest models. With sports-type free-
flight models it is always better to trim for more or less straight
flight or open circles on the glide as this is least likely to show up
spiral-diving tendencies under power, unless the wings or tail-
plane are badly warped.
The high-performance “duration” model—glider, rubber, or
power—normally needs some method of controlling the flight
duration, otherwise the model may readily fly out of sight under
thermal conditions and be lost. (All free-flight models should
carry a label with the name and address of the owner, in any
case, so that the finder can get in touch with the owner if the
model does fly away.)
Such a device is called a dethermalizer, the simplest and most
With a power model the technique is essentially the same—
efficient type being the tip-tailplane (Fig. 77). The mounting of
first trimming out the glide and then concentrating on power-
the tailplane is so arranged that one band holds it in place on
on trim via side-thrust and down-thrust. The main difficulty
its mount whilst another smaller band holds the rear end
here is in getting a “reduced power” climb for the final glide
adjustments. Motors do not “throttle down” easily, although
some can be adjusted to run at reduced power. A timer should
also be used (or a graduated fuel tank) to restrict the motor
run to a matter of 5 to 10 seconds. Failing any other satisfactory
method of “throttling” the motor, merely fit the propeller on
back to front and adjust the engine for normal running. This
will considerably reduce the thrust.
Any adjustment for turn must be approached very carefully
with both rubber and power models. What appears a com¬
paratively gentle turn in gliding flight can wind the model into
120 AEROMODELLING
TRIMMING AND FLYING 121

TABLE XVIII. securely down. A fuse is slipped through this rear band and lit
TRIMMING CHART FOR FREE-FLIGHT MODELS just before launching the model. The fuse burns at a consistent
rate (usually about an inch per minute) and so its action is
Fault Causes Remedy
timed by the length of fuse used. When it burns down to the
Model noses up, Over-elevation
rubber band it breaks the band, allowing the forward band to
Add weight to the nose, which is usually the
then dives. simplest cure. If this is not readily possible, pull the tailplane up to an exaggerated angle (about 35 degrees).
insert packing under the leading edge of the tail-
plane or trailing edge of the wing. As soon as this happens the model will descend almost verti¬
If the wing position is adjustable, move the
wing back.
cally and at a fair speed, sufficient to bring it down from any
Model dives
height through thermals, but not so high that it is damaged on
U nder-elevated Move the wing forward, if adjustable. Other¬
instead of gliding wise, add packing under the trailing edge of the
landing.
properly. tailplane or leading edge of the wing.
There are many other types of dethermalizers, some operated
Model turns (i) Too much rudder tab
violently to one offset
(i) Reduce the amount of adjustment. If the by fuse and others by mechanical timers. The fuse-operated tip
model seems very sensitive, use tailplane
siBe. tilt for turn trim. tail dethermalizer is, however, the simplest and probably the
(ii) Warps (ii) Check wing, tail, and fin for warps. Correct,
if necessary, by heating in front of an most effective type, and certainly the most popular. It is suit¬
(iii) Incorrect alignment
electric fire and twisting straight.
(iii) Check that wing and tail are square with
able for all types and sizes of contest models.
fuselage and each other; also that the fin is
truly vertical and not angled to one side.

Model stalls Over-elevated If glide trim is correct, add downthrust to cure

under power. power-on stall.

Model turns to (i) Excessive turn trim (i) Probably too much rudder tab offset, so
one side and reduce.
dives in, under (ii) Warps (ii) Correct as above. Warps will show up more
power. under power trim than on the glide.
(iii) Too much sidethrust (iii) Only a little sidethrust can be used on a
model without danger of it pulling the
model into a spiral dive.
(iv) Too much downthrust (iv) If sidethrust is used, less downthrust is
used with sidethrust required.

Model does not (i) Lack of power (i) If flight is slow, but steady, almost certainly
climb properly. more power is needed.
(ii) Wrong propeller (ii) Check that you are using the recommended
propeller for the engine.

Model loops (i) Excessive power (i) Engine is too powerful for the size of model.
under power. It may be controllable with downthrust
and sidethrust.
(ii) Incorrect power— (ii) Add downthrust for straight climb; or
on trim downthrust and sidethrust for spiral climb.

Model will not (i) Lack of power (i) Needs a smoother surface to take off from.
take off properly. (ii) Wheels binding (ii) Check that wheels are free-running and
true.
(iii) Undercarriage badly (iii) Try bending the undercarriage back
positioned slightly.

Motor cuts too (i) Wrong adjustment (i) Needle valve adjusted for too lean a
soon. mixture.
(Power models (ii) Poor tank position (ii) Move tank to a more favourable position
only) which does not “starve” the engine in
climbing attitude.
 ENGINES 123

components, and both types also develop far more power for a
given size than a spark-ignition motor. The glow motor, in
fact, can be regarded as a simplified and much improved type
of spark-ignition motor. The diesel is a quite distinct type,
CHAPTER 12
developed independently.
Both the diesel and the glow engine appeared shortly after
ENGINES
World War II and within a period of a few years had completely
replaced the spark-ignition motor as a model aircraft power
There are two basic types of miniature internal combustion
unit. Britain and Continental Europe concentrated on the
engines—diesels and glow motors. The diesel is a compression-
ignition engine which is entirely self-contained and requires
only “diesel” fuel to operate. The glow motor has a special type
of plug (a glow plug) which has to be connected to a battery
for starting, after which the battery can be disconnected and
the plug will remain hot enough to ignite the fuel on each
revolution of the engine. The fuel used in this case is an alcohol
mixture (methanol plus castor oil or synthetic lubricating oil).
Both types have their advantages and disadvantages. Glow
motors are usually lighter and faster revving, and also somewhat
easier to start in the very small sizes. Diesels generally develop
more power for the same size of engine, but have distinct
limitations in the very small, or very large sizes. Thus glow
motors are generally to be preferred in “baby” sizes up to about
0-5 c.c. or slightly greater; and in engine sizes above 3-5 c.c. Fig. 78. The baby size (0.5 c.c. capacity) diesel makes a light, compact
The higher speed of the glow motor is also preferred for power unit for small models, but can be a little tricky to start.

“racing” or certain types of contest models, using special fuels.

diesel and the United States on the glow motor, largely to the
The fact that the glow motor is lighter and vibrates less than a
exclusion of the other type in each case. Only within the last
diesel when running also makes it a preferred choice for radio
few years has the glow motor become a popular type in this
control models. A further advantage in this latter application
country, and then mainly in the smaller sizes up to o-8 c.c.
is that the glow motor is more readily adaptable to multi¬
capacity, and in larger sizes for radio-control models and large
speed control than a diesel via a throttle.
control-line models. As far as production is concerned, British
There is also another type of engine which has a miniature
manufacturers still concentrate on diesels, and American manu¬
spark plug and coil ignition and runs on a petrol/oil mixture in
facturers exclusively on glow motors. The bulk of the glow
a similar manner to a two-stroke moped or scooter engine, etc.
motors used in Europe, in fact, are imported from the United
This was, in fact, the original type of model aircraft engine,
States. The only country which currently has a large production
although very few examples are made today. This is because
of both types is Japan.
both the diesel and glow motor show considerable advantages
The fact that diesels are essentially European in origin and
over spark-ignition. Both types, for example, eliminate the
glow motors essentially American accounts for the fact that
need for a conventional ignition circuit and the weight of such
122 diesel sizes are quoted in c.c. (cubic centimetres) capacity; and

*
 ENGINES
124 AEROMODELLING 125

normally speak of a glow engine size in terms of its c.c. equiva¬

glow-motor sizes in cu. in. (cubic inches) capacity. There are
lent, however, always in cu. in. The figure is also quoted directly
also certain “standard” sizes, established largely by competition
as a size—e.g. -19 and not 0-19 cubic inches. A diesel engine
classes or popular requirements. In the case of diesels these
size, however, is normally spoken of as so many c.c.s.
embrace 0-5 c.c. (now largely obsolete and replaced by glow
To a large extent diesel and glow motors are interchangeable,
motors in this size range), o-8 c.c., 1 c.c., 1-5 c.c., 2-5 c.c. and
as far as a particular model application is concerned. However,
it does not follow that the “equivalent size” of engine is the
TABLE XIX. STANDARD ENGINE SIZES same. Thus a free-flight sports model designed, say, for a o-8
c.c. diesel might be underpowered if fitted with an equivalent

TABLE XX. EQUIVALENT CAPACITIES

Fig. 79. The glow motor is better suited for very small sizes and is made
C.C. i 2 3 4 5 6 7 8 9 IO
down to 0-2 c.c. capacity. A battery is needed for starting.
Cu. in. 0-061 0*122 0-183 0-244 °'3°5 0-366 0-427 0-488 0-549 o-6io
size (-049) glow motor. Also certain sizes or types of models
may be specifically designed around a particular type of motor.
Cu. IN. 0*1 0*2 0-3 0-4 o-5 o-6 0-7 o-8 0-9 1*0 Thus the small power model designed around a -02 glow
C.C. 1-64 3-28 8-19 13-11 16-39
motor might be quite unsuited to powering by an 0-5 c.c.
4'92 6-55 9-83 11 '47 1475
diesel since the latter will be both too powerful and too heavy.
To clarify this question of possible interchangeability, Table
3-5 c.c. In the case of glow motors the popular basic sizes are XXI has been drawn up as a general guide.
•049, -09, -19 and -29—all referring to the capacity in In the case of contest models, of course, the normal choice is
cu. in. Additional sizes which have appeared, are “baby” an engine of the maximum size permitted under the contest
sizes under -049, such as •02 and even -oi; the -15 as rules, when the choice of engine is influenced by the perfor¬
matching the International contest class size of 2 -5 c.c.; the mance of individual engines available rather than type. For
•35 as being more powerful than the -29 for radio-control sport-flying choice is usually dictated by personal preference,
models and control-line stunt or combat; and even larger unless the size or type of model specifically favours a glow
sizes for radio-control flying, such as the -49 up to -6o engine. Modellers who have experience of one type also tend
(approximately 10 c.c.). Table XIX gives a comparison of to stick to it. Thus the British modellers who have used diesels
these different standard sizes. Table XX can be used to con¬ and grown thoroughly familiar with their characteristics
vert c.c. directly into cu. in. or cu. in. into c.c. One does not generally regard them as easier to start and handle than glow
126 AEROMODELLING ENGINES 127
TABLE XXI. familiar with the starting technique and response to adjust¬
APPROXIMATE “EQUIVALENT” ENGINE SIZES ments of controls on a particular engine and for this purpose
the engine should be mounted on a bench stand or similar rig
Sports Type Racing and not in the model.
Diesel Glow Glow
Manufacturers’ instructions are usually specific as regards
Motor Motor
starting and should be followed exactly. In the case of diesels,
there are two controls—a screw on the top of the cylinder
0*5 c.c. •049 —
which adjusts the compression ratio and a needle valve which
oB c.c. — •049
I C.C. •09 — controls the flow of fuel entering the simple “mixing chamber”
i-5 c-c- — •09
2-5 c.c. •19 •15
3-5 c.c. •23 to -29 •19 (

(5 c.c.)1 •29 •29

(6 c.c.)1 •35 •35
(10 c.c.)1 •60 •49

1 Diesels are not made in these sizes.

I
motors. The average American modeller will have exactly the
opposite view. This is purely a matter of familiarity, but it
does also serve to illustrate that there is a considerable dif¬
ference in the general handling characteristics of the two types.
Most new engines require a certain amount of running-in to
wear off initial stiffness. This is particularly true in the case of
Fig. 81. This “049” (o-8 c.c.) glow motor features an induction tube at the
diesels, which are usually assembled with closer clearances than rear of the crankcase and a reed valve, specially suitably for high-speed
glow motors. Running-in also provides an opportunity to get running. For the same size, glow motors must run much faster than diesels
to deliver comparable power.

which takes the place of the carburettor on a larger i.c. engine.

The “mixing chamber” merely comprises a tube through which
air is sucked by rotation of the engine, with fuel fed into the
1
throat of the tube via a small hole in a spray-bar crossing the
throat. The needle valve merely controls the amount of fuel
which can be sucked out of the spray-bar hole at a time.
To fire, both the amount of mixture and the compression
setting must be more or less right. It is thus necessary to
approach starting and control adjustment methodically. A
Fig. 8o. The i c.c. or i -5 c.c. diesel is about the most popular size for sports
basic technique is to open the needle valve rather more than
models (free-flight and control-line), and probably the easiest of all engines the normal running adjustment (about two to three turns is
to handle. Spring starters may be fitted, but are not usual on diesels. typical) and to prime a cold engine by squirting a few drops of
 128
AEROMODELLING ENGINES 129

JontrnireCtIy,Eintouth^ Cylinder trough the exhaust. Mixture connected to the glow plug using a proper clip, when the
rpn, Can tben be snored for the moment, and we can con engine can be flipped over for starting. It should start im¬
to belncreasecfuntil'th1^1'655^011 adJustmenti Compression needs mediately, or within two or three flips. If not, something is
e increased until the engine fires and starts to run but never wrong—so look for the cause as continued flicking over may
IfCred S° that the Cngine becomes stiff to ’turn over only make things worse.
attemotTof’ back °ff the compression before proceeding. To By far the most common cause of failure to start in the case
*“TP* t0 ,for1ce ,the engme over against excess compression of a glow motor is a weak battery. You need a large size of dry
ot a hydraulic lock caused by too much fuel in the cylinder can cell with a i -5-volt glow plug. Small batteries just do not have
w- ln permanent damage to the engine,
tfceil, CXCess comPression the engine may start to run and
Aen labour to a standstill. With lack of compression the enrine
the r &e’ -°r perhaPs only fire occasionally. Once running
the compression can be adjusted to keep it running and the
mixture control then adjusted for fastest running Further
adjustment of compression may then be called for to get
moothest running. Mis-firing indicates lack of compression
Labouring indicates too much com pression. The Tn/shoZ
be adjusted to run on the leanest mixture (the nSle valve
being screwed m as far as possible without causingthe enriOe
amou°nPt nfOUg ^ Starvation) and tbcre should be the least
mount of compression necessary to eliminate “missing ” Fig. 82. For “racing” or “contest” performance diesels usually employ
Having established these adjustments, the needle valve can lie ball race main bearings for the crankshaft and rear rotary valve induction.
opened up (unscrewed) slightly to give a slightly richer mix!
enough capacity and will be flattened in less than a minute if
Shiv Th^ SOnTCVh? thC Cngine Start« to slow or™ n
and for'fi^ h-11 be best setting both for bench running left continually connected. In the case of an accumulator and
and for final adjustment before launching when the engine if a 2-volt glow plug, be sure that the accumulator is fully charged
to start with. Provided the starter battery is man enough for the
,he m0de1' The* *8My rich Sure
job, refusal to start can then only be due to lack of fuel or too
rat ^ * >oaasr«£rPs much fuel—and you can usually judge which by whether the
cylinder seems to be “dry” or excessively “wet.” There is, of
A glow engine has only one control—the needle valve_but course, also the chance that the glow-plug element may have
S1™ P'-g has to be connected to a battery been burnt out. This can readily be checked by removing the
used with" P UgS, “re,?!fiSned fcr -5 *> must on y S plug (or complete head in the case of an integral element),
connecting to the battery and seeing whether it glows or not.
Running—and starting—characteristics of a glow motor will
also be influenced by the fuel used. The design of a glow motor
runninra'SmentTnd'the^'enginTir-thoked” by't"0™31 is matched to a particular fuel proportion and it will always run
over with the intake tube blocked by a r &TZ best on a “recommended” fuel. However, weather conditions
pnmed through the exhaust port. Thetattery^uStat also affect fuel performance, so what is a “recommended” and
9
130 AEROMODELLING ENGINES
131
suitable fuel for, say, southern California, is not necessarily the acquired only through practical experience when concen¬
best for a colder climate. trating on racing performance.
A conventional “straight” glow fuel consists of methanol and For the general run of requirements, however, the basic rule
castor oil in the proportions of about 713. Engine performance is to use the cheapest fuel which suits the engine. That is to say,
can be improved by adding dopes, such as nitromethane. if the engine starts more easily and runs more sweetly on a
Besides increasing the speed and power developed, nitro¬ lightly “doped” fuel, then it will be best to adopt that fuel as
methane also improves starting and promotes smoother run¬ standard. There is no point, however, in going to a more
ning. Only a little nitromethane is needed to give smoother heavily doped fuel (with increased expense) as there is unlikely

Fig. 83. Induction for this rear-rotary diesel is controlled by a port in a

rotating disc inside the crankcase passing the end opening of the intake
tube. The compression screw also has a separate locking device to hold its
setting at high speeds which could produce vibration.
Fig. 84. This twin-cylinder 2-5 c.c. glow engine has front rotary
(crankshaft) induction, with the fuel mixture drawn into the crankcase
running, but performance will go on increasing with increasing through the centre of a hollow crankshaft.
nitromethane content, provided the design of the engine is adjusted
or developed around such a fuel mixture. to be any marked gain in performance unless (i) the engine is
Nitromethane is a very expensive additive and materially designed to match “high-nitro” fuels and (ii) the engine is
increases the overall cost of the fuel. Sports engines, therefore, being used for contest work (e.g. free-flight power duration).
are usually designed to run on “straight” glow fuel or fuels with Diesels are far less fussy as regards fuels. Also there is no
not more than about 5 per cent nitromethane, or equivalent question of “hotting up” a diesel fuel with dope. There are no
additive. Racing glow engines, on the other hand, may be such additives which give a marked improvement in perfor¬
designed to run on fuels containing 50 per cent or more of mance although some, such as amyl nitrite and amyl nitrate,
nitromethane, when the fuel cost may be ten times that of a have a smoothing effect on running. These are often incorporated
“straight” fuel. Because of the altered proportions (mainly in proportions up to about 4 per cent in standard diesel fuels
lowered compression ratio), such engines may not run con¬ for just this effect. Larger proportions do not add any greater
sistently on straight fuels, although some will. Fuel proportions smoothness of running and may even be harmful. Nitro-
may also have to be adjusted to maintain top performance benzine is also sometimes added to diesel fuel to promote better
under different weather conditions, so the whole business economy of running (that is, to reduce fuel consumption), this
becomes somewhat tricky and demands specialized knowledge being particularly significant in the case of team racers.
132 AEROMODELLING
ENGINES
133
A “basic” diesel mixture comprises equal parts of ether, flown in a park or small field. Also, of course, there are likely
paraffin, and lubricating oil. The ether content is a poor fuel to be even more complaints when a noisy engine is “run-in”
and is essential only to promote easy firing under compression. at home or in the garage.
Small diesels tend to be a little tricky as regards “optimum” Silencers can be fitted to most types and sizes of model
proportions of ether and may require this to be increased to engines and have become an obligatory fitting for contest
40 per cent. In other formulas the oil content may be reduced models from January 1965. The fitting of a silencer may result
slightly to increase the amount of paraffin, but never below in some loss of power, but this is seldom more than about 5 to
about 25 per cent for a new engine, or below 20 per cent on a 10 per cent and is negligible in the case of sports engines. As
yet, however, relatively few engine manufacturers produce
silencers specifically for their range of models and these have
to be purchased as an accessory. The established modellers, too,
do not take too kindly towards the fitting of silencers; but the
fact remains that silenced engines do make possible the flying
of models in areas where otherwise the noise would soon lead
to a ban on the operation of such models.

Fig. 85. Rear induction is used for this 2-5 c.c. “racing” glow motor, with
mixture flow controlled by a reed valve.

well-run-in engine. Diesel-engine manufacturers usually specify

a particular fuel mixture which is best suited to their particular
engine but most diesels will run more or less equally well on
any standard diesel mixture, although starting and running
adjustments may differ from fuel to fuel. One of the most
important points is always to keep diesel fuel in closed containers
so that the highly volatile ether content cannot evaporate off.
Stale fuel which has lost much of its original ether content may
make starting very difficult, or even impossible.
It is a characteristic of all model engines that they are
extremely noisy for their size and the faster and more powerful
the engine the noisier it usually is. This has frequently led to
complaints from nearby householders when power models are
operated in built-up areas—e.g. control-line models being
 CONTROL-LINE MODELS
135

in a circular path it must also pull outwards all the time. This
will occur naturally provided the balance point of the model
comes in front of the pivot point for the bell-crank, the farther
forward the balance point the stronger the outward “pull”
CHAPTER 13 produced. At the same time, however, the manoeuvrability of
the model is reduced by such a trim. Thus whilst the typical
CONTROL-LINE MODELS sport or speed model may be balanced well forward, a stunt
model needs to be balanced nearer to the pivot point.
The principle of control-line flying is shown in Fig. 86, the
model being tethered by two lines attached to a handle held by
the “pilot” or operator. These lines also terminate on a bell-
crank securely mounted on a pivot bolt in the fuselage (or

Fig. 87

In this case additional measures may have to be taken to

ensure that the model maintains the lines taut during flight.
There are other factors to consider, too. The stunt model is
capable of wing-overs, loops, etc., where the flight path may
carry the model overhead. There will thus be a tendency for
the weight of the model to act against any outward “pull,”
which again may cause slackening of the lines and loss of con¬
sometimes to the wings in the case of small models). A stiff" trol. This is particularly likely to happen if the flying speed of
wire rod then connects the bell-crank to a horn attached to a the model is fairly low and the line length quite long.
hinged elevator. Thus a rocking movement of the handle will In order to improve manoeuvrability by moving the balance
move the elevators up or down, providing control over the up- point aft towards the bell-crank pivot and still maintain ade¬
and-down motion of the model, or movement in pitch as it is quate “pull” on the lines such methods as engine offset and
technically called. rudder offset may be used, both tending to pull the model out¬
The basic requirement to maintain this control is that the wards during flight (Fig. 87). In addition balance weight may
lines remain taut, i.e. although the model is constrained to fly be added to the outer tip to offset the weight of the lines (which
134
136 AEROMODELLING
CONTROL-LINE MODELS
J37
could cause the model to roll inwards during certain man¬ ploys fairly straightforward, robust construction with moderate
oeuvres) ; or the inboard wing panel may be increased in span rather than high power. Manoeuvrability and line stability
and thus area to provide extra lifting surface for a similar are likely to be limited so that flying is normally restricted to
effect. Line stability is also helped by angling the line backwards shorter lengths and the flight potential limited to climbs and
slightly. dives, wing-overs, and possibly loops. Many models of this type
Thus whilst the fact that a control-line model is flown under are available as ready-to-fly productions moulded in plastic.
control means that free-flight stability problems do not arise, A wide range is also available in the form of prefabricated kits
the more one attempts to improve the aerobatic performance
of such a model the more critical the design requirements
become. Apart from “line stability” or maintaining the lines
taut in all flight attitudes, wing-loading also has to be kept
down to a figure similar to that for free-flight models, calling
for built-up wings and lighter fuselage construction than can
be employed successfully on sports or speed control-line models.
Control-line model size is limited only by available power
units. The same size of control-line model requires more power
to fly than its free-flight counterpart. In practice, this simply
means that control-line models are smaller for similar engine
power, compared with free flight. The maximum length of line
which can be used successfully is also directly dependent on the
model size and engine power. Thus the smallest practical
NO DIHEDRAL
control-line models employing “-049” engine power may be
suitable for a maximum line-length of about 20 ft. (although Fig. 88

shorter lines would be preferable in windy weather), whilst

the largest sizes powered by “35” glow engines may safely which are suitable for assembly by the novice builder. They are
accommodate line-lengths of 60 to 70 ft. an excellent starting point for the beginner to learn the art of
Except for competition work, actual choice of line-length, control before advancing to a more specialized design with a
within limits, is largely arbitrary. Shorter lines can be used to greater performance potential. Unlike the case with a free-
fly in a restricted space, but if the lines are too short the flight flight model, the responsibility for the safety of a control-line
isless “realistic” and the “pilot” has to turn round more rapidly model lies in the hands of the pilot all the time it is in flight—
to follow the model. Long lines can be an embarrassment if the thus learning to fly control-line is an essential requirement for
weather is windy since gusts may result in loss of control success. Fortunately the basic technique is readily picked up
through blowing the model inwards—when the only recourse with a little practice and the use of a robust “trainer” for this
is to step back smartly to take up line tension again and regain stage avoids the damage that could result to a lighter, more
control. It is always safer, therefore, to fly on shorter line- fragile stunt or scale model crashed by mistakes in control
lengths rather than “maximum” lengths. Contest models are movement.
always flown on specific line-lengths, according to class and Having mastered the basic flying technique, the art of stunt
size (see Chapter 16). flying must then be learnt via a proper stunt model. A useful
The sports-type control-line model (Fig. 88) generally em- fact here is that a small stunt model is far less likely to be
 I38 AEROMODELLING
CONTROL-LINE MODELS I39

damaged by a crash than a large model, so it makes a more

The main design difference between a stunt model and a
practical “stunt trainer.” However, unless very carefully de¬
sports model can be summarized as: (i) larger wing area and
signed—and not too small—its aerobatic performance may be
lighter loading; (ii) shorter tail moment arm to improve
limited. For example, although the model may be of proper
manoeuvrability and particularly looping radius or “turning”
“stunt” proportions, if too heavily loaded or underpowered, it
radius in the looping plane; (ii) thick, symmetrical wing
may not perform tight loops. Thus a single loop may be quite
sections for greater wing efficiency and good performance in
hazardous and consecutive loops out of the question. The good
inverted flight; (iv) particular attention to design features
maintaining line tension during manoeuvres.
Although stunt models all tend to look rather similar, there
are two distant lines of approach—the “minimum-size” model

with a rather higher wing loading and faster flying speed, and
the large area model which has a “free-ffight” loading and
flies more slowly. The former is typical of the diesel-powered
stunt control-line model, particularly in the smaller sizes
matched to engines of up to 2 -5 c.c. capacity. The larger stunt
models are almost invariably powered by “29” or “35” glow
motors with wing areas ranging from 400 to 600 sq. in.
The “combat” model is really another form of stunt model,
stunt model, on the other hand, will perform loops with height
except that it is intended to be flown two or more in the same
to spare, leading to consecutive loops, figure eights, and so on.
circuit on identical line-lengths—each model flown by its own
A good performance in inverted flight is also essential in a sound
pilot. Each model tows a streamer and the object of combat
stunt model design calling for the use of a fairly thick sym¬
flying is for each pilot to attempt to attack and cut the
metrical wing section (Fig. 89).
streamer(s) of his opponent(s). Whilst a thrilling and spec¬
The typical stunt model tends to be larger than its sports
tacular sport, the risk of collision or loss of control is con¬
counterpart, with considerably more wing area and an ele¬
siderable and crashes are frequent. As a consequence the combat
vator movement of 30 to 45 degrees up and down. To improve
model is built strongly, although the “write-off” rate still
manoeuvrability elevator movement may also be combined with remains high! To utilize tougher and heavier structures, and
wing flap movement in the opposite direction (i.e. flaps lowered still keep down to the lighter loading requirements for good
as elevators are moved up, and vice versa) (Fig. 90). A special
manoeuvrability, models of $his type are usually of flying-wing
“stunt” tank will also be necessary for feeding the engine with
layout, with the elevator hinged directly to the trailing edge
a constant supply of fuel, regardless of the flight attitude.
(Fig. 91). Otherwise, the design requirements are basically the
 140 AEROMODELLING CONTROL-LINE MODELS 141

same as for stunt models. The combat model can also be flown shown in Fig. 92. Originally there were no limiting specifi¬
individually as a “stunt” model, if preferred. cations regarding model proportions and official classes were
Another popular “contest” type is the team racer. These are merely designed by engine size. More recently minimum
based on specific class sizes, governed by engine capacity (see areas have been specified as a requirement of the model design.
Chapter 16) and flown on standard line-lengths. Two or more Since the overall requirement is purely maximum speed,
models are flown in the same circuit, the object being to cover basic requirements boil down to a streamlined model of
a given distance (number of laps) in the minimum time. Pit minimum size based around a suitable engine (the most power¬
ful available in its class!) and propeller. Relatively little control

is needed—only enough to be able to take off and control the

stops are usually necessary for refuelling, calling on close co¬ model in level flight, and bring it in to a smooth belly landing.
operation between the pilot and “pitman” who refuels the The undercarriage is invariably omitted to save drag, the
model and restarts the engine after each stop. The models model taking off from a wheeled “dolly” which it drops or
themselves are invariably of semi-scale design, with cowled in leaves behind as soon as it is airborne (although speed models
engines, although some concession to “model performance” is may also be hand-launched). Wing sections are made thin
often seen in the use of single wheel undercarriages. (again to reduce drag) and the whole model is often polished
A team racer can also make an attractive “sports” model for or burnished to a superfine finish. The greatest proportion of
the individual flyer who is attracted by “realism.” Design pro¬ drag is then accounted for by the lines.
portions are such, however, that manoeuvrability is restricted To reduce line-drag, line-diameter is kept to a minimum
to climbs, dives, and wing-overs. Equally, sports control-line (limited only by strength requirements). “Monoline” con¬
models although not conforming to team-race contest speci¬ trol may also replace conventional two-line systems, again to
fications can be used for “team racing” by individual groups of save the drag of one wire. In this case control is realized by
modellers. applying twist or torsion to the single line, this movement
Speed models are essentially models for the specialist and being translated at the model end as bell-crank movement via
they all tend to have a similar overall appearance, like that a screw drive. Again this is strictly for the expert modeller
 142 AEROMODELLING CONTROL-LINE MODELS
*43
TABLE XXII. since the piloting technique required is quite different and has
TYPICAL SIZES OF SPORTS MODELS AND TRAINERS to be re-learnt.
Engine Size
Ultimately the performance of the speed model depends on
the engine/propeller combination, top performances normally
0 5 c.c. o-8 c.c. i-i 5 c.c. 2 5 c.c. 3’5 c.c. being realized by extensively reworked and specially tuned
Diesel or Diesel Diesel Diesel Diesel or
•049 Glow
“racing” engines. The average modeller has virtually no chance
•29 Glow
of success in competition against the speed control-line experts,
Span 15-18 in. 20 in. 24 in. 30 in. 36 in. many of whom (especially on the Continent) are sponsored by
Wing chord 3i in. 4~4i in. 5 in. 6 in. 7 in.
Wing area sq. in. 60-70 80-90 120
the engine manufacturers. Engine performance is also directly
180 250
Propeller sizes 6x4 6x4 7x4 8x6 9x8 related to a suitable fuel tank installation, the flexible “bladder”
5X4 6x6 7x6 8x8 10x6 tank being highly favoured since pressure feed is more reliable
9x6
Line length 12-18 ft. 15-22 ft. 20-30 ft.
than conventional tanking. Even the flying of a speed model
30-40 ft. 50-60 ft.
Lines Thread 33 s.w.g. 30 s.w.g. 30 s.w.g. Stranded itself is highly specialized, competition requirements calling
wire or wire wire or c/1 wire for the use of a special pivoted yoke mounted on a pylon in the
thread stranded
centre of the flight circle, in which the pilot must rest his arm
c/1 wire
so as to avoid any possibility of assisting the performance of the
model by “whipping.”
Control-line models undoubtedly offer the would-be model-
TABLE XXIII.
designer the greatest scope since stability problems are reduced
TYPICAL PROPELLER SIZES FOR STUNT MODELS
to a minimum. Control-line is also by far the safest solution for
Engine Size flying scale models, particularly for flying for fun, and where a
large amount of detail work, etc., can be incorporated without
1-5 c.c. 2 -5 c.c. 3 5 c.c. 5 C.C. 0*35 cu. in.
having to worry too much about added weight, or the possi¬
(o-igcu.in.) (o-2gcu.in.)
bility of serious damage resulting through lack of “free flight”
Prop, diameter 8 7 9 10 10 stability. Choice of prototype is not limited to single-engined
Prop, pitch 4 6
8
4 or 0
nr9
4
c or 10
4 or 6 4 5 5 or 6 aircraft for twin-, three-, and four-engined flying scale (con¬
trol-line) models are perfectly practical. Added realism can be
introduced by operating retractable undercarriages, wing-
TABLE XXIV. TYPICAL SIZES OF SPEED MODELS flaps, bomb-dropping, etc., triggered by a third line (normally
slack but pulled to trigger the operation of a secondary control).
Engine Size
Such devices are not practical on free-flight models, except
(o-8 c.c.) (2-5 C.C.) (3-25 c.c.) (5 c.c.) (10 c.c.) with elaborate (and expensive) radio-control systems.
•049 Glow •15 Glow • 19 Glow •29 Glow •60 Glow

Span 11-12 in. 14 in. 15 in. 17-18 in. 20 in.

Wing area (sq. in.) H-15 24-28 40-50
25-30 30-35
Tail area (sq. in.) 6 9-10 10 18
15
Weight (ounces) 4-5 8-10 9-10 12-14 20-24
Length 9 13 18
14 '5
Lines 36 s.w.g. 33 s.w.g. 33 s.w.g. 30 s.w.g. 28 s.w.g.
 RADIO-CONTROLLED MODELS
145
battery and actuator. The radio link, however, enables “on-
off” switching to be accomplished over a distance of several
miles, if necessary (although the normal ground-to-ground
CHAPTER 14
range of typical model radio-control systems is about half a
mile).
The simplest form of actuator is the escapement, which itself
RADIO-CONTROLLED MODELS
is powered by a rubber loop (Fig. 93). A radio signal received
The remote control of free-flight models via a radio link has
been developed to the stage where the radio equipment avail¬
able is extremely reliable and requires no specialized knowledge
of electronics either to install or to operate it. All such equip¬
ment is designed to operate within a frequency band of 26-06
!5La£2p ™efacyciles/second, this waveband being allocated by
the t-r.F.O. for model radio-control systems. A G.P.O. licence
is required to operate any such equipment, the cost being Si
for a five-year period. A licence is obtainable on application,
no techmcal or other qualifications being required.
The basic elements involved in any model radio-control
system are a transmitter, a receiver, and an actuator. The
Fig. 93
transmitter is invariably designed to transmit a constant fre¬
quency carrier signal within the permitted frequency band and
by the receiver is passed on to the actuator either directly
is commonly crystal-controlled to ensure a stable signal fre¬
(relayless receiver) or indirectly (through closure of relay
quency (although this is not obligatory in this country). The
contacts) to cause current to flow through the actuator coil.
receiver may be designed to respond directly to the carrier
This pulls in the armature, allowing the escapement to rotate
signal, or to a lower frequency tone signal superimposed on the
a quarter of a turn under the action of the rubber motor, when
transmitter carrier. The latter is normally the preferred system.
it is brought up against another stop. This quarter-turn
In either case the receiver response takes the form of a change
movement can be used to move a control surface via suitable
m current flowing through the receiver circuit. This in turn is
linkage. On release of the radio signal the current through the
used to operate a relay whose contacts then form an “on-off”
coil ceases, the armature drops out and the escapement rotates
switch for the actuator circuit. Alternatively, the relay may be
a further quarter turn. At the same time the attached linkage
dispensed with and the receiver current amplified to a suitable
returns the control surface to its original or “neutral” position.
level to operate the actuator direct (relayless receiver).
The next signal, producing a further quarter-turn rotation, will
ihe actuator is an electro-magnetic device providing the
move the control surface in the opposite direction, returning to
power to move a control. As far as the mechanical working of
neutral again on release of signal. Such an actuator, therefore,
the complete system is concerned the transmitter-receiver com¬
provides alternate control positions in sequence, returning to
bination merely acts as an “on-off” switch for the actuator.
neutral (self-neutralizing) on release of signal.
The same effect would be produced by using a manually-
A transmitter, receiver, and actuator of this type form the
operated switch m the actuator circuit comprising simply the
I44 basis ofa simple single-channel system, which isa perfectly practical
146 AEROMODELLING
RADIO-CONTROLLED MODELS 147
method of controlling a free-flight model when the control
Receiver, batteries, and actuator are located in the centre of
action is linked to the rudder. It is possible to provide additional
the fuselage to concentrate these weights around the balance
controls, again in sequence, from a single signal or single point of the model. The fuel-tank fitted is larger than normal in
radio channel by designing the escapement with a greater
a free-flight model, usually containing sufficient fuel for flights
number of “stopping” points. This, however, causes practical of 10 or 15 minutes duration.
difficulties both in signalling correctly to the required sequence
The main limitation with single-channel radio is lack of
position and as regards speed in selecting a particular control
elevator control. Various ingenious systems have been devised
position. The usual limit is a “third” position on the actuator
whereby both rudder and elevators can be operated by a
which can be signalled at will to operate a second actuator,
single control channel and one, known as the “Galloping
which in turn then operates a further control (usually engine
speed). An alternative arrangement is to use this “third”
position to hold an elevator either “up” or “down” for a
change in trim.
Owing to the development of transistor circuitry which, in
addition to compact sizes for the receiver, means that battery
requirements are very modest, the complete installed weight of
a lightweight single-channel radio-control system can be as
low as 3 ounces. Thus radio control is a practical proposition
for free-flight power models from about 20 in. span upwards.
Whilst ideal for operating in small spaces, such tiny models
are, however, suitable only for calm-air flying. For training Ghost” actually provides proportional movement of rudder or
purposes a 40-48 in. single-channel model is much better, elevator. All such systems, however, have their distinct limi¬
powered by a 1 to 1 -5 c.c. diesel or glow motor equivalent. This tations. The only fully satisfactory method of extending control
gives scope for providing both rudder and engine speed control coverage is by utilizing multi-channel radio control. Unfortunately
(the latter via a second actuator coupled to an engine with this also greatly increases the cost of the radio equipment.
throttle control), using a “compound” actuator. Engine speed The conventional multi-channel radio system operates on
control—either “fast” or “slow,” selected in sequence by tone signalling, but instead of a single tone superimposed on the
signalling the “third” control position on the actuator, provides transmitter carrier separate “tone” switches on the transmitter
a means of bringing the model down as well as controlling enable a number of different tones to be signalled indepen¬
direction by the rudder. dently. Each tone then provides a separate channel for control
The cost of a complete single-channel radio-control system signalling. The receiver, in turn, must be capable of decoding
can be as low as £10 where the transmitter and receiver are the various tones and responding by a switching action as far
assembled from kits, or roughly twice this figure using finished as the associated actuators are concerned.
commercial equipment. At best, however, single-channel radio There are two basic methods of “decoding” multi-channel
only provides partial control and models have to be designed tones. The first, and simplest, uses a reed bank which is something
with about the same amount of inherent stability as free-flight like a relay with a number of individual reeds replacing the
sports models, although the proportions and fuselage outline armature. Each reed is trimmed to a specific length and reso¬
shape may differ. The usual—and most successful—layout is nates or vibrates when a particular tone signal is fed to the
the high-wing monoplane with semi-scale lines—Fig. 94. reed bank coil. The vibrating reed then closes a corresponding
 AEROMODELLING RADIO-CONTROLLED MODELS
149
actuator circuit connected to the reed contacts (Fig. 95). The to a full control position, and also to provide automatic
reed bank simply has the same number of resonant length drive back to the neutral position when the tone signal is
reeds as there are transmitter tones to be decoded. Each reed released—a self-centring multi-servo. Alternatively, the self¬
circuit completed by its contact when the reed is vibrating then centring action can be eliminated so that the servo can be
either operates a relay switching the actuator, or supplies “inched” in either direction by appropriate tone signals—
current to operate the actuator direct through a suitable progressive multi-servo. Self-centring servos are used for the
amplifier circuit to boost the current (relayless receiver). main flying controls (rudder, elevators, and ailerons); and

RUDDER'

a-CHARREL RECEIVER

REED ELEVATORS
BARK

ERCIRE -
THROTTLE

A/LERORS

DIAGRAMMATIC ORLY

The other basic method of decoding the transmitter tones is progressive servos for trim controls (elevator trim, aileron or
to employ electronic filters in the receiver circuit, each filter rudder trim, and engine speed) (Fig. 96).
passing only one particular tone to operate its specific relay. The particular virtue of multi-channel signalling where two
This leads to a more complicated (and expensive) receiver, and channels control one servo is that a particular control position
also a heavier unit because of the additional filter components is signalled direct, that is, there is no lag in having to switch
involved. It does, however, have the advantage that the indi¬ through a sequence to repeat a control movement, as with
vidual tone frequencies can be more widely separated and thus single-channel actuators. Also, of course, given enough chan¬
there is less risk of interference between adjacent tones. nels, direct control can be provided on all necessary control
The actuator in the case of a multi-channel system becomes surfaces.
a servo signalled by two tones, each servo being associated with For complete control of a free-flight power model eight chan¬
a particular control. The servo itself is, basically, an electric nels are required, covering rudder, elevators, ailerons, and
motor. Operation of one tone signals the motor to drive in one engine speed. A further control for elevator trim is also de¬
direction, and operation of the second tone signals the motor to sirable, making ten channels required in all. A model with this
drive in the opposite direction. Built-in switching is incor¬ control complement can be flown under control all the time in
porated in the servo unit to stop the motor when it has driven a similar manner to a full-size aircraft. Provided the “pilot”
150 AEROMODELLING RADIO-CONTROLLED MODELS
!5i
has the necessary skill, therefore, the model does not have to Anything less than eight channels cannot give complete
be stable itself. It is, in fact, a decided advantage for the model control, so in such cases a different type of design is required
to have more or less neutral stability, when it will be that much with a certain amount of normal free-flight stability. Usually
more manoeuvrable. It must not be unstable, however, as this this means a high- or shoulder-wing layout, with a more
could over-ride the control available to correct, e.g. if an un¬ generous dihedral angle on the wings. Rudder is an essential
stable model was deliberately put into a spin it might not be control, which would normally be allied to engine speed (with
possible to pull it out of the spin again by corrective action on four channels) or engine speed and elevators (with six channels).
the controls. With just two channels only rudder control can be used.
In the case of a radio-controlled glider, of course, complete However, this will still show distinct advantages over single¬
control coverage can be given with only six channels (rudder, channel rudder-only control since rudder response is signalled
elevator, and ailerons). However, an additional two channels directly. Multi-channel radio is, in fact, the only real answer to
for elevator trim control will be a distinct advantage. At the satisfactory radio control of aircraft. Single-channel radio
other extreme, four channels (covering rudder and elevators) retains its popularity only because it is so much cheaper.
would provide adequate control for slope soaring, etc., provided The ultimate in multi-channel operation is achieved with
the model had a reasonable degree of free-flight stability. proportional control systems. With conventional radio control
The only real limitation to increasing the number of channels all the main flying controls (rudder, elevator, or ailerons) are
with “multi” radio is one of cost. The modern relayless “multi” either full “on” (control signalled and held) or “off” (control
receiver with a ten-reed bank can be used for two to ten signal released and the servo self-centring to neutral). With
channels, simply by adding servos and their amplifiers. The proportional control, movement of any control surface is
cost, however, goes up by a matter of about £10 per servo directly proportional to the movement of the corresponding
(i.e. £10 for every additional control service added). The control stick on the transmitter, in just the same manner as the
additional weight and bulk is seldom critical, for the individual controls of a full-size aircraft follow exactly the movement of the
servos can be quite small in size and weigh little more than control column. Flying a model aircraft with proportional
2 ounces each. radio control is, therefore, exactly the same as flying a full-size
Typical multi-channel aircraft designs have features which aircraft, except that the pilot is standing on the ground rather
would make them definitely unsuitable for free flight (without than sitting in the aircraft. This does present some difficulties
control). The low-wing layout is widely favoured, with only in co-ordination but, once the technique is mastered, the con¬
moderate dihedral angles and moderate, rather than exag¬ trol achieved is considerably smoother than that possible with
gerated, tail-surface areas. Complete control enables them to be conventional “multi” controls.
flown safely in high winds or other conditions which would The main drawback is that the additional complication to
“ground” the normal free-flight sports model; also the power the radio side, and the actuators, considerably increase the cost
used is higher than that for normal free flight. Given a reason¬ of such equipment. Thus whilst the cost of a complete ten-
able design layout, in fact, much of the success of a multi¬ channel conventional multi installation may be £ 100 to £150,
channel aircraft depends on its having enough power to per¬ the cost of a comparable proportional system is at least doubled.
form manoeuvres in the vertical plane. Full power does not This puts it well out of the range of the average model en¬
have to be used all the time for, with throttle control, the engine thusiast.
speed can be adjusted as required. As a consequence “49” and Although the majority of modern radio-control equipment
even “60” glow motors are commonplace fitted to models of is extremely reliable, there are still snags, notably on the ques¬
5-6 ft. wingspan. tion of interference. The conventional receiver circuit is of the
 RADIO-CONTROLLED MODELS
153

super-regenerative type which is highly sensitive but not

selective. That is to say, although tuned to a particular trans¬
mitter it will almost certainly respond to any spurious signals
which might be present within the model waveband (e.g.
another transmitter). A super-regen receiver may also interfere
with another receiver fairly close to it. It is thus impossible to
fly more than one model at a time with a super-regen receiver
as any other transmitter operated within a mile or so is likely
to cause interference, although not directly tuned to that
receiver. This applies whether the transmitter is crystal-
controlled or not. The average super-regen receiver is suffi¬

Rudder, engine, elevators,

ciently non-selective and broad in tuning to pick up any trans¬

Rudder, engine, elevators

mitter signal over the full range of 26.96 to 27.28 mc/s.
•% « g
W > g The solution here is to employ a superhet circuit for the

Rudder, engine, trip

Oh <U ^5
.et-tj
^ D oj
w receiver, although again this increases the cost. The particular
gҤ s virtue of the superhet is that the tuning is extremely sharp and,
with crystal control, it can be matched exactly to a “spot” fre¬
§ § « u quency established by the transmitter crystal. Another trans-
U«J U
U O UV UV cw u tT tT uT «i

elevator

ailerons
u v u t) e mitter-superhet receiver combination tuned to a different
13 13 "O 13 o
■O T3 ? -n T) >i 13 13 13 13
3 3^33" 3 3 33 3 JJ “spot” frequency within the permitted band can then be
04 Pi 13 Pi p4 '3 P4 Pi '3 operated simultaneously without interference. In fact, as many
as ten or twelve transmitter-receiver combinations may be

8- or io-channel
u
"u "u ® operated simultaneously at different “spot” frequencies be¬
Single-channel

g e §
g s s tween 26.96 and 27*28 mc/s, although five or six is a more
2 33
5 —1 o
6-channel
-C u V usual number.
y y c o
ii«“ The additional cost of the superhet receiver is thus justified by
»|l 3 o
c C o its freedom from interference, whether or not it is intended to
CO X to 00
operate simultaneously with another modeller. It is not com¬
pletely free from interference. A “broad” (i.e. non-crystal
controlled) transmitter signal could interfere with it by over¬
2 riT °°p
iO V
CO lapping the spot frequency; or any other spurious signal em¬
bracing the spot frequency. It is, however, very much better
than the super-regen receiver in this respect and it is also
capable of responding to weaker transmitter signals because of
its greater selectivity, so that its range tends to be improved
with the same transmitter. In other functional respects it is
similar to the super-regen receiver and is applicable to both
single- and multi-channel units.
 INDOOR MODELS *55
Extreme patience and skill are needed both to construct and
to handle models of this type, which are quite outside the scope
of the average model-builder’s abilities. Smaller models of this
type may, however, be produced quite successfully for small-
CHAPTER 15 space flying with duration capabilities of a minute or more; and
similar stronger and heavier free-flight models may be suitable
INDOOR MODELS for tissue covering with similar scope. Conventional built-up
*Models M tissue-covered rubber-powered models are either too large or
suitable for flying indoors embrace a variety of dif¬ too heavy, and fly too fast for satisfactory indoor flying.
ferent types, ranging from small rubber-powered “round-the- Basic principles of the “round-the-pole” model are shown in
pole” or tethered models for small-space flying to highly
specialized ultra-lightweight duration models capable of flight
durations of more than half an hour and requiring the use of
an airship hangar or similar large, unobstructed building for
flying!
Performance with all indoor models, free-flight or tethered, is
entirely dependent on length of motor run (except that chuck-
gliders may be flown in large enough buildings). The i.c.
engine is largely ruled out as a practical power unit for indoor
models, which means that rubber is the standard motive
power. Motor run is then primarily dependent on being able
to use the smallest possible motor cross-section (for maximum
number of turns per inch of motor length) together with the
largest possible diameter propeller. In turn, this means an
ultra-lightweight model. Thus the 30-minute-plus indoor dura¬
tion model may have a span of some 30-36 in. With a propeller Fig. 97. The model is tethered by means of a very light line to a
of 16 in. diameter powered by a two strand motor of 1/16 swivel fitting on the top of a short pole and flies in circles round
sq. in. cross-section, the complete model weighs a matter of the pole for as long as the rubber motor develops enough thrust
1/20 ounce! via the propeller. With very lightweight construction, and
To get down to such extreme lightness the construction is microfilm-covered wings and tail, rubber section may be re¬
very fragile, involving the use of carefully selected balsa of duced to the extent that 5-6-minute flight durations are pos¬
minimum density and “sparless” wing and tailplane construc¬ sible on a line length of some 12 ft. and a pole height of 6 ft.
tion, covered with microfilm. The propeller is also normally Tissue-covered models require a more powerful motor because
built up as an ultra-light balsa frame, and is also microfilm- of the increased weight, but still may be capable of flights of
covered. Microfilm itself is formed by pouring a few drops of 3 minutes’ duration or more, with specialized designs and
clear dope (suitably plasticized) on to water, where it spreads skilful construction. More robust models, flown purely for fun,
out into an extremely thin film and sets. The film is then lifted may achieve flight durations for from 30 seconds to a minute or
off by means of a wire hoop, applied to the framework as a more, flying much faster and with smaller propellers. Many of
covering and the edges are trimmed off with a hot wire. these are based on the smaller rubber-powered flying-scale
«54
 156 AEROMODE LLING INDOOR MODELS I57
kits, fitted with much larger propellers to prolong the motor run. 18 in. and a line-length of 5 ft. 6 in., with the official distance
Round-the-pole or “RTP” flying also lends itself to “speed” for timing purposes ten laps.
models, designed expressly with powerful rubber motors and An essential requirement for satisfactory free-flight perfor¬
small high-rewing propellers, the motor run being sufficient mance with any indoor model is a very low flying speed, which
to cover a specified number of laps at maximum possible can only be achieved by employing the lightest possible struc¬
speed (Fig. 98). A longer “course” can be covered by “pit ture coupled with a reasonable size of model. It is not just a
stoppages” for rewinding the motor, the winner being deter¬ question of model size. Thus a small “outdoor” rubber-
mined as the one with the least elapsed time to complete the powered model is not suitable for indoor flying, even in a
relatively large hall, because it flies too fast and is very prone
to suffer damage on striking an obstacle. The true indoor free-
flight model flies literally at less than walking pace, with the
large diameter propeller turning over very slowly. Attempts to
make the model smaller whilst retaining a similar form of
construction inevitably result in a loss of performance since
certain material weights cannot be reduced in direct proportion
to size. The result is a higher wing loading, and a greater
flying speed.
The 12-15 in. span microfilm-covered free-flight model can,
however, be trimmed to fly in small circles and give flights of a
minute or more in an average-size room—flight duration
normally terminating when the model strikes a wall and slides
down it (the model will not weigh enough to crash!). In a larger
hall similar models may produce flight durations of several
whole distance (number of laps). Speed models of this type have
minutes under conditions too restricted to operate a larger
a popular appeal for junior aeromodelling club activities
model. The small (15 in.) model may well weigh more than its
during the winter months when outdoor flying is restricted by
larger counterpart, and thus be less tricky to construct and
the weather. RTP duration flying, which once had a wide
handle. It is an interesting field for the serious enthusiast to
club and National contest appeal, is no longer particularly
investigate, particularly since models of this type cost very little
popular, although it is recognized as a “National” contest type.
to build and can be “sized” to suit the flying area available.
Two “official” sizes are specified—Class A for models up to
Every school, for example, will have a hall large enough to
1 ounce maximum weight, and Class B for models between
provide space for microfilm model flights of several minutes’
x and 2 ounces in weight. Pole-heights are 3 ft. and 6 ft. re¬
duration, and perhaps for similar performances with very
spectively; and line-lengths 6 ft. and 12 ft., respectively. This
lighdy-constructed tissue-covered models.
general rule that the line-length should be twice the pole-
The best durations are invariably realized with a “stick”
height applies for all “duration” RTP flying, whether for fun
model layout, since this is by far the lightest and simplest type
or contest work. In the former case the line length can be selec¬
to construct. The fuselage consists of a light, hollow stick made
ted to suit the space available, e.g. a sitting-room.
by rolling thin balsa sheet into the form of a tube or “teardrop”
For speed RTP models pole-height is usually made shorter section. This stick carries the rubber motor suspended beneath
in proportion, the official S.M.A.E. size being a pole-height of it. Tailplane and fin and lightweight structures are carried on a
I5a AEROMODELLING INDOOR MODELS 159
separate boom attached to the stick fuselage, whilst the wing is not so light as for “duration” models since flight performance
invariably “parasoP’-mounted on thin struts (Fig. 99). These will be limited, to start with anyway. The scope offered by
struts plug into tiny tubes cemented to the top of the stick fuse¬ indoor models is, however, far more restricted and less satis¬
lage and also provide a means of adjusting the incidence of the fying than that provided by outdoor models, unless one finds a
wings to trim, if necessary. particular attraction in the challenge offered by the micro¬
A similar basic layout is used in all sizes of such models, the film-covered “duration” model.
real secret of weight reduction being in the choice of balsa and

very light frames 40% OF

MICROFILM-COVERED WING AREA

APPROX, */s SPAN

LIGHT TAIL BOOM

HOLLOW BALSA
TUBE

POLYHEDRAL WING

LARGE DIAMETER
MICROFILM-COVERED
PROPELLER

the use of minimum material sizes consistent with just enough

rigidity in the complete structure. In the case of larger models
the wings, and often the fuselage stick too, are braced with
very thin tungsten wire, involving very intricate construction
and painstaking workmanship. It is virtually only possible to
“build down to weight” through a whole series of models,
starting with more generous wood sizes and using thinner,
lighter stock as experience is gained. Attempting to duplicate
a record-breaking model for a first attempt is almost certainly
foredoomed to failure.
A reasonable size of indoor flying space also gives scope for
trying unorthodox models, such as ornithopters and small,
simple helicopters. Construction again must be kept light, but
 CONTEST MODELS l6l
in Chapter 1, grouped broadly under “free-flight” and “con¬
trol-line,” with specific types and classes in each category. In
addition a number of general rules apply to models flown in
contests, with which entrants must be familiar. Such regula¬
CHAPTER l6 tions are summarized in a Contest Rule Book published by the
S.M.A.E.
CONTEST MODELS
Gliders
The International sport of model flying is run under a Sporting
Code issued by the Federation Aeronautique Internationale The World Championship formula for model gliders is
(F.A.I.), who are also responsible for the ratification of World known as the A/2 and calls for a model conforming to the
Records. Annual or bi-annual World Championship events in following specification—
various categories are also held under F.A.I. regulations. (i) Total area of wings and horizontal tail surface to be
National contests and national organization are handled by between 496 and 527 sq. in.
individual bodies in the countries concerned, usually affiliated (ii) Minimum total weight of the model to be 14-46 ounces.
to or part of the National Aero Club. The British authority is (iii) Maximum loading (i.e. weight divided by total area) to
the Society of Model Aeronautical Engineers (S.M.A.E.), who be 16-38 ounces per sq. ft.
are responsible for ratifying National Records, organizing The standard method of launching is by a tow-line not
National contests, selecting British teams, etc. The general exceeding 164 ft. in length.
sporting and contest activities in the country are organized by (Note: The odd and fractional specification figures result
local clubs, themselves affiliated to the S.M.A.E., with their from the fact that world championship specifications are
own local flying grounds. These in turn may organize flying derived originally from metric units.)
meetings, normally called galas, which whilst not having Gliders of other size or type may, of course, be used for
National status as competitions attract contestants from other “open” competitions organized on a National or club basis.
clubs all over the country. The A/2 merely represents a specialized contest class, and the
A particular virtue of local club membership—apart from only one recognized for World Championship events.
the good-fellowship of meeting and flying with other aero-
modelling enthusiasts—is that third-party insurance coverage Rubber Models
is provided, although this can also be negotiated by individuals
who are not club members. Insurance cover of this nature is The World Championship class here is the “Wakefield”
formula with the following specification—
inexpensive and is thoroughly to be recommended to all
(i) Total area of wings and horizontal tail surface to be
model-flyers, especially anyone who operates a power model.
As regards overall definitions, a model aeroplane is defined between 263-5 and 304-5 sq. in.
as having a total surface area (combined area of wings and (ii) Minimum total weight of model to be 8 • 11 ounces.
horizontal tailplane) of not more than 16-14 sq. ft., and not (iii) Maximum loading to be 16-38 ounces per sq. ft.
exceeding 11 -023 pounds (5 kilograms) in weight. In the case (iv) Maximum weight of rubber motor to be 1 -768 ounces.
This is a “restricted” formula in the sense that the limitation
of power models, motors should not exceed 10 c.c. capacity
to rubber weight considerably reduces the potential perfor¬
(o-6i cu. in.).
mance of a model which could be built within the specification
The various categories of model aeroplanes are as described
160
 l62 AEROMODELLING

items (i) and (ii). For this reason many rubber models are built
for higher performance (unrestricted rubber weight) for “open”
competitions.

Power Models

For World Championship and International events the

maximum size of engine is restricted to 2.5 c.c. The following
restrictions then apply to the model design—
(i) Minimum total weight of model to be 10.58 ounces per
c.c. of engine capacity.
(ii) Minimum loading to be 6-55 ounces per sq. ft.
(iii) Maximum loading to be 16-38 ounces per sq. ft.
(iv) Maximum duration of engine run to be 10 seconds from
release of the model.
National (British) contests also admit of unrestricted power
models where only the engine-run restriction above applies;
and the |A power class where engine size is restricted to a
maximum of 0-85 c.c.
There is also a Pay-Load category where the model is re¬
quired to carry a dummy pilot of specified dimensions and
weight in an acceptable pilot’s compartment in the fuselage.
This again is a special “restricted” formula. For engine sizes
up to 0-82 c.c. the model (less fuel and payload) must weigh
at least 5 ounces; and for engines from 0.85 to 1 c.c., 6 ounces.
The standard pilot” must weigh 4 ounces, with an additional
payload of 1 ounce in the case of motors up to 0.82 c.c., and
2 ounces for engines larger than 0-82 c.c.

Control Line Stunt (aerobatic)

No limitations are placed on model design or construction

other than the restriction that the model must not weigh more
than 6 pounds and the engine must not be larger than 10 c.c.
Lines must be of steel, and any length between 25 ft. and
70 ft. can be used, with the following minimum line diameters
specified—
Engines up to 2 -5 c.c. 33 s.w.g.
» » 2-5-6-0 c.c. 30 s.w.g.
C__ _ r*
33 33 6-10 c.c. 26 s.w.g.
 164 AEROMODELLING

The complete control system must also be capable of with¬

standing a pull of at least ten times the weight of the model
Scoring points are allocated for the quality of performance
of a sequence of standard manoeuvres.

Control-Line Team-Racers

Three classes are recognized, based on engine size and with

model dimensions, etc, related accordingly. These specifica¬

TABLE XXVII. CONTROL-LINE SPEED CLASSES

tions are summarized in Table XXVI. The line-length in the
case of Class JA is equivalent to 18 laps to one mile; and in
Class B to 14 laps to one mile. The line-length in the case of the
International Class A is equivalent to 10 laps equalling one
kilometre distance covered.
Standard distances run are 5 and 10 miles in Class JA and
Class B; and 10 and 20 kilometres in the Class A. The lower
distance m each case corresponds to heats and the longer
distance to finals. 6

Control-Line Speed

One World Championship formula and five additional

classes are recognized in this category. The World Champion¬
ship formula is of a “restricted” type, permitting the use of
one of two standard fuels only (80/20 or 75/25 methanol/
castor), thus excluding diesel engines. Maximum engine size is
limited to 2-5 c.c. and the model must have a minimum total
area (wings and horizontal tail surfaces) of 2 sq. decimetres
(31 sq. in.) per c.c. of engine displacement; and a maximum
loading of 3276 ounces per sq. ft. Radius of the flight circle is
52 It. 2 ? m, giving 10 laps equal to one kilometre. Minimum
me diameter permitted is 0.0098 in. (33 s.w.g.) in the case of
two lines; or 0-0136 in. (29 s.w.g.) in the case of monoline
control.
Details of the other five speed control-line classes are sum¬
marized in Table XXVII for convenience of reference. All
speed flymg for official timing is done with an anti-whip pylon
(see Chapter 13).
166 AEROMODE LLING

Control-Line Combat INDEX

There are no restrictions on model design or construction but A2 GLIDERS, 38
Ducted fan, 72 et seq.
Actuator, 144
for the purpose of S.M.A.E. (National) competition engine size Adhesives, 26
-models, 74
Duration design, 49
is restricted to a maximum of 3-5 c.c. with line-length stan¬ Adjusting diesels, 128
Aileron control, 67 Electric power, 85
dardized as 50 ft. The streamer carried by the model should be All-balsa models, 29 et seq. Engine capacities, 124
10 ft. long by 11 in. wide, securely attached to the rear of the Anti-warp structures, 92 Equivalent engine capacities, 124
Augmenters, 69 -sizes, 126
fuselage or fin by a 5 ft. length of strong thread. The total Autogiro, 79 Escapement, 145 ,
period of combat time is 5 minutes, with points deducted at the Automatic pilot, 64 Expanded polystyrene, 25
Auto-rudders, 42
rate of 1 for every 15 seconds the model is not airborne during Fan types, 73, 74
Balance point, 111 Fin construction, 95
this time; add points scored at the rate of 5 for each single Balsa cement, 26 Finishing solid balsa, 108
cut of an opponent’s streamer. — cuts, 22 Flaps, 139
— grades, 20 Flex wing, 81
— selection, 21 Formers, 89
— weight, 21 Free flight, 9, 59 et seq.
Bamboo, 23 -stability, 11
Battery (glow motor), 129 Fuel proofing, 108
Bell-crank, 134 Fuselage construction, 87 et seq.
Best glide trim, 115
Geodetic, 92
Breaking in rubber, 55 Glass fibre, 25
Building board, 17
Gliders (definition), 10
Canard, 83 Glide trim, 113
Carving a propeller, 57 Glow fuels, 130
Catapult launch, 43 — motors, 122
Centre of gravity, 111
Half shell fuselages, 89
Chuck gliders, 29
Hand launching gliders, 31
Colour doping, 107
Hardwoods, 22
Combat models, 14, 139
Helicopters, 77 et seq.
Complete control coverage, 149
Compression adjustment, 128 Indoor free flight, 157
Contest specifications, 161 Initial trim, 113
Control-line design, 135 Jets (definition), 11
-flying, 134 Jetex charges, 69
-models, 9, 14 — design, 71
-model sizes, 136, 142 — helicopter, 78
-Speed Classes, 165 — installation, 70
-stability, 135 — model sizes, 71
— movements, 137 — unit, 68 et seq.
Corded motors, 54
Covering, 102 et seq. Kit models, 25
— materials, 26, 109 Kits, 15
— sheet balsa, 108 Line drag, 141
— tips, 104 Longerons, 97
Cutting ply, 19 Longitudinal dihedral, 112
Delta wing, 84
Making up rubber motors, 54
Dethermalizer, ng Material sizes, 97 et seq.
— fuse, 121
Maximum turns for rubber motors, 56
Diesel, 120
Microfilm, 154
— fuels, 131
Minimum flying speed, 115
Dihedral, 11
— sinking speed, 115
— joints, 90 Modelling knife, 17, 19
Doping, 107 Monoline, 141
Doped fuels, 131 Multi-channel controls, 149
Downthrust, 117 — systems, 147

X67
 168 INDEX

Needle valve, 127 Scale sailplanes, 37

New engines, 126 Semi-scale glider, 38
Nitromethane, 130 -models, 12
Nylon covering, 107 -power, 61, 62
Obeche, 22
-power sizes, 65
Original design, 13 -rubber model, 50
Omithopter, 80 Semi-solid construction, 32
Pendulum controls, 64 Servo, 148
Perspex, 24 Shaped charges, 69
Plastic cement, 26 — sections, 94
Pliers, 18, 20 Sheet balsa construction, 33
Plywood, 23 — fuselages, 33
Ply thicknesses, 24 — plastic, 24
Polyhedral, 60 — wings, 34
Power duration design, 60 Shoulder wing, 62
— helicopter, 77 Sidethrust, 117
— models (definition), 10 Silencers, 33
— model sizes, 65 Silk covering, 107
— trim, 116 Single-channel systems, 145 et seq.
Pretensioned motors, 54 Solid construction, 29
Prime, 127 Speed models, 15, 141
Proportional controls, 151 Sports models, 14
Pulse jet, 75 Spray-bar, 127
Pushers, 82 et seq. Spruce, 22
PVA glue, 26 Standard engine sizes, 124
Pylon wing, 60 Starter battery, 129
Starting diesels, 127
QuARTER-grain balsa, 21, 22
— engines, 127
Racing engines, 142 — glow motors, 128
Radio-control frequencies, 144 Streamlined fuselages, 89
-licence, 144 Stunt models, 14
Radio-controlled models, 152 — model design, 138
Razor blade, 19 -propeller sizes, 142
— saw, 19 Superhet receiver, 153
Reed bank, 147 Super-regen receiver, 153
— cane, 23
Relayless receiver, 148 Tail-less models, 83 et seq.
Ribs, 91 et seq. Tailplane construction, 93 et seq.
Rigging, m — tilt, 118
— details, 114 Tautening tissue, 105
Round-the-pole flying, 154 Team Racer, 14, 140
RTP model classes, 156 -specifications, 163
— speed, 156 Tethered flying, 154
Rubber helicopter, 77 Third-line controls, 143
— lubricant, 55 Tissue covering, 102
— model construction, 49 Tools, 18 et seq.
— models (definition), 10 Tow-hook positions, 41
— model design, 47 Tow launching, 40
-sizes, 51 Tow-line glider design, 37
-sizes, 46 — stability, 44
-types, 47 et seq. Trimming chart, 120
— motors, 52 Turn trim, 118
— motor sizes, 53 Types of models, 9
Rudder control, 66 Undercambered wings, 104
Running in, 126
Wakefield formula, 46, 47
Saucer wing, 85 Warren girder, 92
Scale models, 11 Water-energized batteries, 86
— modifications, 11, 63 Winding rubber motors, 58
— power, 63 Wing construction, 90
 . wide variety
IPer whetting his
appetite on one . ■_ i. Is the urg° to m’e
both cars and track for him:
Electric Model { rs provi*>- 'ould-be car builder with the
expert guidance mat shd'ijTd im (or her—there is already
quite a following of women “o. y to build and run model cars
that would be at least as good as anything that can be bought, and,
with only moderate care and a capacity for taking pains, will be very
much better. Only the simplest tools and materials are needed;
very little space is required since the finished car is unlikely to be
more than five inches long. Indeed, the whole “workshop” can go
in a small attache-case.
Detailed instructions are given for making racing, sports, and
“drag” cars, including both chassis and body; the book also covers
track building, trackside scenery, speed-controlling mechanisms,
timing devices, and gives a simple description of basic principles,
so no one need hesitate to take these first steps to being a “World
Champion” in miniature.
The author has been keenly interested in electric model cars
since they first appeared in the magazine which he was editing some
ten years ago. He has continued to encourage the development of
this branch of modelling until today it is part of an immense industry,
and he retains his interest and activity by editing Model Cars, a
magazine solely devoted to the hobby.

Illustrated, 21/- net

MUSEUM PRESS
26 Old Brompton Road London SW7