Dynamic

Modeling
by Stan Hall, EAA 10883
1530 Belleville Way
Sunnyvale, CA 94087

tion and flown by experts. Its behavior

is judged by persons having a well-de-
veloped sense of what it is trying to tell
them. In short, the model is an en-
gineering tool, designed, Grafted and
used like the precision instrument it is.
It should be recognized early on that
even very accurately scaled models do
not represent true, miniature analogs of
full scale when compared on the basis
of performance, as will be seen as this
article develops. However, much can be
learned from them and, if one takes the
test data derived therefrom with a pinch
of salt, much of it can be extended with
modest validity to full size.
Models can be particularly useful
where the design departs significantly
from what we have come to consider
"conventional" and/or the flight behavior
of the model turns out to be gross, rad-
ical or erratic. In both cases, significant
clues to full size behavior are offered.
(Taylor and Holcomb state that they
learned a great deal from the fact that
their model crashed. That can be con-
sidered a solid data point!) However,
such clues can be considered valid only
if the model is dynamically similar to
the aircraft being simulated. This is to
say that the propulsive power, the
weight and the distribution of that
weight are all in scale with full size. A
model built only to linear scale and little
Use of Free-Flight, Dynamically-Similar Models else may be considered simply a recre-
ational model; having limited use as an
In Estimating Full Scale Aircraft Behavior aid to full-scale design.
The thoughtful experimenter is drawn
to consider what the "big boys" (the
major manufacturers) are doing, or not
The January 1987 issue of Sport As defined here, a dynamically simi- doing, with free flight, dynamically simi-
Aviation carried a provocative article lar model is one whose size, propulsive lar models. On the one hand it is prob-
by Molt Taylor and Jerry Holcomb on power, weight and weight distribution ably safe to say that if such models
the use of free-flight, dynamically-simi- are all in scale with the full size aircraft were as useful to full scale design as
lar models in estimating certain impor- being simulated. It is a model which, we would suppose, the majors would
tant full-scale parameters by way of like its full size counterpart but unlike a be making extensive use of them. Such
simulation. recreational model, responds to inertial does not, however, appear to be the
It is the purpose of this article to ex- as well as aerodynamic forces. The ob- case.
pand on the principles so ably intro- jective is to have it fly in scale with its On the other hand, unlike homebuil-
duced by Taylor and Holcomb. It pro- full size counterpart. ders such as Taylor and Holcomb, no-
vides numerical scaling factors and re- The model is assumed to be non-in- body ever accused the manufacturers
marks appropriate to designing and strumented and radio controlled. Flight of personal airplanes, at least, of being
building the model, flying it and sub- test data are taken by eyeball, a stop- particularly innovative. But they do
sequently extending the test data de- watch and maybe a tape measure. The have computers which (of course) solve
rived therefrom to full scale. model is built to a high degree of perfec- all the problems.
30 JULY 1987
 inversely as the cube of X. The weight
Fig. 1 - Example Calculation for Pitching Moment of Inertia of Full Size Aircraft of the model would, then, be 2645
pounds divided by 8 cubed, or 5.16
Note: This calculation is shown only to illustrate technique. A complete calculation would pounds. A model having a span of 4.48
show scores of weight items in the table below instead of only 5. feet (35.8/8) might be difficult to build
on a budget of only 5.16 pounds.
The situation is made even more
pressing by the requirement (explained
later) that only about half of this weight
be in the model itself, the remaining half
being represented by the installation of
movable ballast weights. These ballast
weights are needed to permit adjust-
ment of the model's Moment of Inertia,
more of which later.
Going from 1/8 scale to, say, 1/5
scale makes a large difference in model
weight and thus tends to make building
it easier and more accurate. A 1 /5 scale
model of the Cessna 172 would weigh
Wt. (w), Ibs. Mass, slugs* just over 21 pounds, half of this (10.16
250 7.76 Ibs.) going into the model itself and half
100 3.11
500 15.53 3.0 into ballast. Building a model with a
80 2.48 span of 7.16 feet (35.8/5) for 10.6
50 1.55 pounds doesn't seem outside the limits
8.2 of practicability. If it turns out to be so
in a specific case, the model should be
I = 1149.7 + 8.2 =1157.9 slug ft.2 made even larger.
'Mass = w/32.2 Another reason for making the model
large relates to Reynolds Number (RN),
which is based partially on wing chord.
Measuring "Scale" In A Dynamic the original one. As explained later, so long as the
Model Weight varying as the volume repre- Reynolds Number is above 120,000 or
By way of definition, in the kind of sents the dimensional conversion of so the increase in airfoil drag coefficient
recreational model we see flying on weight to linear scale. Dimensional con- with decreasing RN is not likely to be
weekends, "scale" refers, of course, to version is symbolized by the Greek let- serious, although it should be consid-
that fraction of full size to which the ter, Lamda (X) and is the reciprocal of ered when extrapolating model data to
model is built. Size in this case refers linear scale, e.g., in a 1/5 scale model full scale. Similarly with the decrement
to linear dimensions such as span, X would equal 5. in maximum lift coefficient.
length, etc. Dimensional conversions for other The 1/5 scale model Cessna, having
In dynamic models the term "scale" factors to be scaled in a dynamic model a wing chord of just under a foot and
has an additional connotation, depend- appear in Table 1 as "scaling factors". flying at a scaled maximum speed at
ing on the specific factor under consid- Although the other factors shown in the sea level of 64 miles per hour (see
eration. The scale is not linear, but is Table cannot be analyzed as simply as Table 1), would be flying at a wing RN
expressed in terms of linear scale. For was done in the tin can analogy for of about 614,000, which is comfortably
example, although a 1/5 scale model weight, they have been analyzed and above the suggested 120,000
would have a wing span 1/5 as long as tested against full scale aircraft. The minimum. However, the farther re-
its full size counterpart, the scale physics is valid. moved from 120,000 the more accurate
weight would not be 1/5 the weight of The essentials of Table 1 are adapted the extension of model flight data to full
the full size aircraft, but 1/125. This is from a work (Ref. 1) published 37 years scale.
because weight varies as the cube of ago by a brilliant young engineer named A third reason for making the model
linear dimensions. Ernest G. Stout, who in time came to large is that its dynamic behavior in
You can prove this to yourself by con- guide the technical fortunes of the WW flight (pitching, rolling, yawing) is more
sidering the case of an ordinary tin can. II Consolidated-Vultee Aircraft Corpora- quantifiable than in a small model be-
If you double its size you double every- tion (now General Dynamics). Stout in- cause small models tend to be more
thing about it; its diameter, its height troduced aircraft dynamic modeling to "twitchy" in flight which makes eyeball
and the thickness of its "skin". When the U. S. in 1938 and reports having assessment of their behavior more dif-
you double the diameter and height you experienced a high degree of success ficult.
find you have squared the area (multi- with it. A fourth reason for using a large
plied it by four), so if you stop right here model is that measuring its Moment of
you have squared the weight. However, Inertia tends to be more accurate.
since you also double the thickness of Model Size and Weight
the skin you in effect add four times For a number of reasons it is advisa- Model Airfoil Selection and Reynolds
more to the "area". In sum, you no ble to make the model as large as prac- Number
longer have area but volume, which for ticable. First, the larger the model the According to Schmitz (Ref. 2), de-
the double-scale can is now 8 times the less difficult it is to hold the weight to pending upon the chord of the wing and
original volume, bringing 8 times the scale values. the speed of flight, airfoils intended for
weight. 8 is 2 cubed, right? Consider, for example, a 1/8 scale use on full size aircraft suffer a perfor-
Going in the other direction, if you cut model of the Cessna 172, which has a mance loss when used on models; the
the can to half size, the weight scales wing span of 35.8 feet and (in one ver- drag coefficients are higher while the
to 1/2 cubed, which is 0.125, or 1/8. The sion) a gross weight of 2645 pounds. maximum lift coefficients are lower. Just
1/2 scale can now weighs 1/8 that of According to Table 1, weight varies how much the loss in performance is
SPORT AVIATION 31
difficult to say because good test data
on models tend to be in short supply. Fig. 2 - Determining Moments of Inertia via Compound Pendulum
This makes extrapolation of perfor-
mance to full size less than accurate. In Basic Equations (Fits Both Cases Shown Below)
order to ease the problem somewhat,
Schmitz recommends the use of thin-
ner, more highly cambered and w (case 1) = Model wt., Ibs.
sharper-nosed airfoils. w (case 2) = Wt. of bob wt., Ibs.
In reviewing some basic or, solving for T where lcg
( = As shown, ft. Make as long as
aerodynamics, one notes that the main practicable.
is known: T = Period of oscillation, sec.
reason for the difference in airfoil perfor-
mance between model and full size is g = 32.2 ft/sec2
T = 2TT
L = Assume zero for models
because the air flow interacts differently
with the model than with the full size
aircraft. This difference is reflected, of
course, in the familiar term, Reynolds CASE 1 • OVERHEAD PIVOT CASE 2 • PIVOT AT C.G.

Number.
Reynolds Number is an expression PITCHING MODE /-FIXED SUPPORT

which relates the viscous and inertia

forces in the airstream boundary layer.
Numerically, it equates (at sea level) to WIRES
about 800 times the wing chord in in- STIFF WIRES
ches, times the flying speed in miles per
hour. A model having a wing chord of
six inches and flying at 25 miles per
hour is operating at a wing RN of about
120,000.
The reader will note the frequent re-
currence of Reynolds Number in this ar- ROLLING MODE FIXED SUPPORT

ticle. This is because much of the "prob-

lem" with models can be assigned di-
rectly to their characteristically low RNs. STIFF WIRES
The message carried by Reynolds
Number is that some very low drag
coefficients and high maximum lift coef-
ficients can be achieved where the flow
is smooth (laminar) and the flow re-
mains attached to the surface. As is
generally known, however, laminar YAWING MODE
flows tend to be very unstable, breaking
FIXED SUPPORT
away from the surface with little provo-
cation. And of great significance is the
fact that any time the flow separates
from the surface there results a quick OR
and dramatic increase in drag. Whereas
low RNs encourage laminar separation,
high RNs encourage the flow to remain
attached.
If the flow is turbulent but still at-
tached the drag will be higher than in
attached, laminar flow, but lower than doesn't scale at all at low RNs, and Scaling the Model to Fit the Power
in separated flow. Turbulent flow tends hardly at all at high RNs. Test data are Available
to delay separation. needed for the specific airfoil or shape
Achieving attached laminar flow at involved, measured at the Reynolds As Table 1 shows, power varies as x
RNs below about 100,000 is virtually Number involved. raised to the power of 3.5. The 1 /5 scale
impossible with any practical airfoil con- In order to minimize the effects of low model used as an example in the Table
struction. RN, modelers often purposely force the requires 0.57 horsepower in order to
Some idea of the influence of RN on flow to become turbulent, this by lead- scale with the 160 horsepower of the
drag may be seen in data provided by ing edge trip wires, thin, sharp-nosed full scale aircraft.
Hoerner (Ref. 3). His data show a 12% airfoils or combinations of these and The question arises, what does one
thick streamlined section operating at other strategems. do if a motor of the required power is
an RN of 100,000 to have about 2.5 Models flying at RNs above 120,000 not available? The answer is, one re-
times the zero-lift drag coefficient of the or so show considerably less influence scales the model to whatever pow-
same section operating at 1,000,000 on drag and maximum lift than do mod- erplant is available. Determining the
RN. A 20% section operating at those els operating below that value. For this new scale is simple, but a calculator
same RNs shows a drag coefficient reason it would seem prudent to use capable of handing fractional expo-
close to 4 times higher at the low end wing sections not over 12% thick on nents is needed.
of the RN range than at the high end. models used to simulate full scale air- Using the above case as an example,
Trying to extrapolate model RN data craft, even though the latter might have consider the use of a motor rated at
to full size may be considered an exer- thicker sections. Structural considera- 0.40 horsepower; the closest to 0.57
cise in futility. Drag, which is a prime tions will, of course, bear heavily on the horsepower assumed available. Simply
ingredient in speed performance, decision. divide the full scale power by 0.40 and
32 JULY 1987
raise the result to the power of 1/3.5, or Thus, in this instance the modeler is left Clearly, if the same torque is applied
0.2857. X turns out to be 5.54 instead with the option of ignoring the problem to both flywheels the lighter flywheel will
of 5, and the model scale is no longer or trying to guess the effect — or re- come up to a given rpm more quickly
1/5, but 1/5.54. For quick reference and scaling the whole model to fit the pro- than the heavier one. It will also come
for those users having calculators of li- peller. The author's vote would be to to a stop quicker when the same brak-
mited capability, Table 2 does this job ignore it and hope for the best. How- ing torque is applied.
for selected ratios of full scale power to ever, if the model is so small that the Applying this analogy to airplanes,
model power. propeller RN is really threatening and continuing the absurdity of the
One can use motors as small or as 120,000, the logical option would be to example to make the point, consider
large as desired so long as the model make the model larger. two airplanes of identical dimensions
is scaled to fit. In using small motors, Since reciprocating model engines and, like the flywheels, one of iron and
however, be advised (again) that if the normally turn up much faster in terms the other of balsa. If these aircraft are
model turns out so small that Reynolds of rpm than do full size aircraft engines, rotated in pitch about their respective
Number considerations become signifi- some difficulty will likely be encountered CG's from the same input from the tail
cant, extending flight test data to full in getting the propeller to scale both in (from elevator movement, say), the
scale will suffer. The same can be said rpm and power absorbed. One solution balsa aircraft will respond much faster
of model weight, where as stated ear- to this problem would be to use a speed than the iron one. This is because its
lier, the smaller the weight the more dif- reduction drive; to make the propeller pitching moment of inertia is lower.
ficult to manage. rpm lower than the rpm of the engine. Moment of inertia (or in this case,
It is recognized that since model This would at least make it more ac- mass moment of inertia) is simply the
airplane (reciprocating) engines are ceptable if not solve the problem com- product of the mass of each part of the
normally rated in terms of displacement pletely. Such reduction drives are com- aircraft and the square of its distance
rather than horsepower, some difficulty mercially available on the model mar- from the aircraft CG; the products all
can be extected in relating the two. How ket. being subsequently added together.
to solve this problem is left to the in- Using a reduction drive requires, of (Also refer to the example shown in Fig-
genuity of the reader; displacement sel- course, the design of a propeller capa- ure 1.)
dom correlates with horsepower from ble of absorbing scaled power at less Mass, of course, is simply the weight
one engine to the other. than scale rpm; a larger diameter pro- of the object (for convenience, in
Although knowing the rated horse- peller with, perhaps, non-scale blade pounds) divided by "g" (32.2 feet per
power of the engine is essential, even angles. As is generally known, propeller second squared). To keep the units
more useful would be a curve of full design is a science in itself, one cer- consistent, distance is measured in
throttle horsepower versus rpm be- tainly beyond the scope of this article. feet. The product comes out in units of
cause with this curve one could throttle slug feet squared. Moment of inertia is
the engine to the required output, as- Dynamic Behavior and the Moment symbolized by the letter "I" and, since
suming the engine were big enough in of Inertia the I is taken about the aircraft CG, it is
the first place. In considering the dynamic behavior symbolized by "lcg". When considering
If the experimenter can handle the of the model (pitching, rolling, yawing) pitch, roll and yaw separately there is
weight of batteries, some good electric as a precursor to full scale one needs no confusion in calling any one of them
motors are commercially available. observe that dynamic events occur at 1,-g. When considering them in combina-
Such motors are commonly rated by faster rates in the model than in full tion, the identification has to be
horsepower, or watts, from which horse- scale. However, although practical con- changed in order to keep the bookkeep-
power can be easily derived. siderations might mitigate against in- ing straight.
strumenting the model to determine the It is not normally considered practical
Propeller Scaling dimensions of these events, one can at to compute the lcg of a model, as is
If all the propeller linear dimensions, least approximate the time during almost always the case in full scale
rpm and blade angles are scaled in ac- which they occur. Time can be scaled. airplanes because the model's indi-
cordance with Table 1, the propeller As Table 1 shows, it varies as the vidual parts are too small and light to
helix angle (V/nD) of both the model square root of X. yield anywhere near accurate values for
and full scale propellers will be equal, For example, a 1/9 scale model (X = the leg. And here is where models pay
and so will the power coefficient (Cp). 9) can be expected to pitch close to 3 off; you build the model and determine
Thus, the power absorbed by the pro- times (the square root of 9) as fast as its leg by test. The procedure is
peller will be in scale with full size, as the full-size aircraft for a given control explained later.
will the thrust. input. Stated the other way around, the As stated earlier, the example
However, the influence of the lower full-size airplane will pitch about a third Cessna 172 has a gross weight of 2645
Reynolds Number of the model propel- as fast as a model in this scale. As a pounds and a 1/5 scale model system
ler still needs to be considered because clue, to subsequent full scale behavior should weigh just over 21 pounds.
of the increased blade drag coefficient; in flight test, the test pilot is sure to tuck The model itself should actually
propellers do have small chords. this away in his memory bank. weigh about half this value because
The Reynolds Number of the Cessna In order to make the model pitch, roll you'll be adding identical movable bal-
172 propeller at the three-quarter radius or yaw "in scale", it is vital that its mo- last weights on each side of the CG to
(the usual propeller reference radius) at ment of inertia about the appropriate bring the total weight to 21 pounds with-
2750 rpm and 144 miles per hour is on axes through the CG be in scale, too. out altering the CG position. Using two
the order of 1,250,000. The RN of the This requires, of course, that the mo- weights in this manner takes care of the
1/5 scale model propeller, taken at the ment of inertia of the full size aircraft be "pitch" mode, such weights being suffi-
same radius fraction, at a scaled 6160 known, at least in close approximation, cient if only the pitching behavior is of
rpm (see Table 1) and 64 miles per hour beforehand. interest (frequently the case). If rolling
computes to about 165,000. Although Moment of inertia is a measure of a and yawing behavior (such as maneu-
this is close enough to the recom- rotating body's resistance to accelera- vering, spinning, etc.) are also of con-
mended 120,000 minimum to warrant tion. To illustrate: Consider the rather cern, two additional weights need be
some concern there's not much one can absurd case of two flywheels of identi- disposed equally about the CG, in a
do to raise the RN without unduly com- cal dimensions, one made of iron and spanwise direction. This means that the
plicating the whole scaling exercise. the other of balsa. individual ballast elements will be lighter
SPORT AVIATION 33
than in the case of pitch only, the decre- period per oscillation of 2.74 seconds the title of this article, most of what is
ment in moment of inertia being made as calculated from the equation for T in written here deals with scaling the de-
up by moving the weights farther apart. Figure 2. We can test to see if this 2.74 sign of the model down from full scale.
The reason for doing all this is to per- seconds actually gives us 0.431 slug However, scaling the model's behavior
mit your actually measuring the lcg, and feet squared by introducing the 2.74 up to full scale remains the objective.
altering it later by moving the weights if into the equation shown in Figure 2 for Note that the term "behavior" is em-
needed to reflect the proper scale. lcg. This checks out close enough for all phasized over performance. This is
There are two approaches to testing, practical purposes to the required 0.431 done for good reason; pitching, rolling
each employing the principle of the slug feet squared. and yawing behavior is more confi-
compound pendulum and each giving A small caveat: The term "|0" appears dently extended to full scale than, say,
the same answer. (Also refer to the in the upper equation of Figure 2 and speed, which is one element of perfor-
sketches shown in Figure 2.) again in Figure 1. This term represents mance.
One method involves hanging the the I of the bob weights around its own As the reader might certainly gather
model, say, from a single point in the CG and for precision it should be com- by now, it does not appear feasible to
ceiling of your workshop, on two wires puted. It also represents the I of items extend model speed performance in its
or cords, one well forward of the CG in the model or the airplane about their various parameters to full scale with ac-
and the other well aft. You now have a CG. Since I0 takes into account the curacy. Again the main culprit is
compound pendulum. By giving the shape of the weight item as well as its Reynolds Number. One can only hope
model a small, gentle push in the ap- mass, calculation usually calls for dig- that the data derived from the model
propriate direction and timing its oscilla- ging out the physics texts for an equa- will at least be indicative of full scale
tions, you can determine the oscillatory tion suited to the shape. This can be performance.
period (T), which is simply the total more trouble than it is worth because Schmitz gives one experimental data
number of seconds divided by the total the value of I0 is sure to be miniscule point of interest in this connection; a
number of cycles. (Recall that one cycle compared with the lcg of the model as manned sailplane he examined showed
is one complete swing, to and fro.) Of determined without it; probably less a maximum lift to drag ratio of 20. How-
course, the greater the number of cy- than 1%. Hence, it can usually be neg- ever, the best L/D a 1/10 scale model
cles (should be at least 30) and the lected. In computing the l^ of a full size of the sailplane could generate was 10.
longer the suspension the greater the airplane some accounting for the I0's of Schmitz neither showed a correlating
timing accuracy, which is vital. Small er- large, heavy items such as the engine, scaling factor nor derived one, mainly,
rors in timing beget large errors in lcg. fuel and crew is often taken, even one might suppose, because of the lack
Knowing the period, the weight (w) of though their impact on the result is sure of sufficient data on the effect of
the model and the vertical distance (t) to be small. Reynolds Number in the low RN range
of the CG from the pivot point in the involved.
ceiling, you can calculate the 1^ of the Scaling Factors For Estimating Full If one is prepared to accept model
model, using the upper equation shown Scale Behavior data on these terms, one may proceed
in Figure 2. Contrary to the implication carried by in extending the data to full scale in ac-
Another technique is to make the CG
of the model itelf the pivot point and
complete the compound pendulum by
hanging a bob weight below it on a pair Table 1 - Scaling Factors
of fairly stiff wires or lightweight
(wooden) struts. Needless to say, it is X = Full Scale Linear Dimensions
vital that friction at the pivot point be Model Linear Dimensions
held to a minimum. The actual weight
of the bob is unimportant; maybe 1/4 Model Design
Example for 1 /5 Scale Model (X = 5)
the model weight. You can now go Parameter Model Should Be: Full Scale Model
through the same timing exercise as be- Linear Dimensions Full Scale/\ Span: 35.8 Ft. 35.8/5 = 7. 16 Ft.
fore and from the data thus obtained Area Full Scale/X 2
Wing: 174 [Link]. 174/25 = 6.96 [Link].
calculate the lcg. Volume, Mass, Force Full Scale/\3 Gross Wt. = 2645 Ibs. 2645/125 = 21.1 6 IDS.
It is more than likely that the lcg deter- Moment Full Scale/X4 Full Scale/625
mined from your first test will differ from Moment of Inertia Full Scale/X5 Pitch: 1346slugft.2 1346/3125 = 0.431 slugft.2
the lcg required. In this case move the Linear Velocity Full Scale/ /X Max: 1 44 mph 1 44/2.24 = 64 mph
ballast weights in the model a little Linear Acceleration Same as Full Same as Full Scale
(equidistant from the CG so as to not Angular Acceleration Full Scale x X Full Scale x 5
Angular Velocity Full Scale x/X Full Scale x 2.24
alter its position) and test again. Repeat
this procedure until the required lcg is Time Full Scale//x Full Scale/2.24
Work Full Scale/x" Full Scale/625
obtained. Lock the weights down for
Power Full Scale/X3 5 Rated: 160 hp 160/280 = 0.57 hp
flight. Wing Loading Full Scale/X 15.2psf 15.2/5 = 3.04 psf
Again using the Cessna 172, which Power Loading Full Scale x/X 1 6.5 Ibs./hp 1 6.5 x 2.24 = 37 Ibs./hp
is reported to have an lcg in pitch of Angles Same as Full_ Same as Full Scale
1346 slug feet squared, as an example R.p.m. Full Scale xyx Rated: 2750 rpm 2750x2.24 = 6160 rpm
and noting from Table 1 that moment of
inertia varies as the fifth power of X, the Full Scale Performance from Model Test
pitching moment of inertia we need Example for 1/5 Scale Model (X = 5)
develop in the 1/5 scale model is 1346 Full Scale Measured Derived
Parameter Should Be: Model Pert. Full Scale Pert.
divided by 3125, or 0.431 slug feet Model x 2.24
Time Model x/X
squared. Maximum Speed Model x/X 64 mph 64 x 2.24 = 1 44 mph
To achieve this value let's hang the Max. Climb Rate Model x/)T 344 fpm 344 x 2.24 = 770 fpm
21.16 pound model 6 feet from the ceil- Takeoff Distance Model xX 160ft. 160x5 = 800ft.
ing and start it swinging. We continue, Pitch, Roll & Yaw Rates Model//X 50°/sec. 50/2.24 = 227sec.
timing the oscillations and moving the
ballast weights until we measure a
34 JULY 1987
cordance with Table 1. Consider the fol-
lowing examples:
Table 2 - Factors Used in Scaling Model to Power Available*
Take-Off Distance (For Users Having Calculators of Limited Capability)

Take-off distance is a linear dimen-

2857
sion, of course, and distance is directly Full Scale HP X = f Full Scale HP |
proportional to X. As the lower part of Model HP Avail. 1 Model HP Avail. 1
L
Table 1 shows, if a 1/5 scale model 100 3.73J
200 4.54
were to get off the ground in, say, 160
400 5.54
feet, the full scale aircraft would be ex- 600 6.22
pected to take-off in 160 times 5, or 800 800 6.75
feet. 1000 7.20
This assumes, of course, the ab- 1200 7.58
sence of Reynolds Number effects. 1400 7.92
Such is not precisely true, of course, 1600 8.23
because the low RN's encountered in 1800 8.51
the ground roll represent higher values 2000 8.77
of the drag coefficient, which impact the
take-off acceleration. But in this case 'Interpolation OK
the drag may be considered secondary
in importance to the mass of the aircraft
because during most of the ground roll
the greater part of the propulsive power feet per minute the climb rate of the full scaled, what you see in the model is
is taken up in accelerating the mass up scale aircraft would compute to 344 likely to be what you get in the full size
to take-off speed, while little is used to times the square root of 5, or 770 feet aircraft — in proper scale, of course.
overcome aerodynamic drag. The re- per minute. Note from the Table that Unlike calculating aerodynamic per-
verse is true, of course, once the aircraft angle of climb would remain the same. formance, calculating dynamic behavior
is in flight and climbing out. is a task of monumental dimensions,
As a point of interest, Stout reports Roll Rate one best left to the professionals. In
that a 1/8 scale, dynamically similar support of this advice, pick up any text
Roll rate is an angular velocity which on aircraft dynamics. If you can get past
model of the XP4Y-1 flying boat left the can be expressed in degrees per sec-
water at a speed and in a time (and the first page you have real mathemat-
ond. Table 1 shows that angular veloc- ical talent — and probably make your
thus in a distance) in scale with the full ity in full scale varies inversely with the
size aircraft. However, he appears to living at it. Unfortuantely, insofar as is
square root of X. If a 1/5 scale model known to this author, no books on air-
have "fudged" a bit on the model by in- showed a roll rate of, say, 50 degrees
corporating full-span leading edge craft dynamics aimed specifically at the
per second the full scale aircraft would homebuilder exist.
slots, which the full scale airplane didn't be expected to roll at 50 divided by the
have. Not to worry. If you'll reflect on the
square root of 5, or about 22 degrees proposition that errors in predicting
Thus, although scaling model take-off per second.
performance up to full scale as illus- aerodynamic performance are less
trated is not entirely accurate, some likely to be threatening to life and limb
good clues are offered. Concluding Remarks than errors in predicting pitch, roll and
Parenthetically, determining take-off Although dynamically similar models yaw behavior, you'll stop searching for
distance by calculation alone is often offer clues to the behavior of full scale that non-existent book or that profes-
unrewarding because of the large aircraft, "clues" have different scales of sional and build a model instead. In so
number of variables involved. Calcu- validity. doing you stand a fair chance of beating
lated distances seldom match those As has been suggested here, clues the professional at his own game.
measured in test. It is likely that an ac- to aerodynamic performance (speeds) Finally, a suggestion to those model-
curately scaled model would, in spite of have a lower level of validity than those ers who build solely for competition. For
the reservations just expressed, do a relating to inertial behavior (pitch, roll, a real competition, make your models
better job because most of the variables yaw). not only in linear scale, but in dynamic
are already "in the model" and its envi- The lower credibility of aerodynamic scale as well.
rons, and the model knows it — proba- clues is not seen as a serious obstacle
bly better than the computer does. to full scale design, however, because References
Stout's model, by the way, had a calculating estimated performance in 1. Stout, Ernest G. — "Development
span of about 14 feet and weighed full scale is no longer the complicated of High-Speed Water-Based Aircraft",
close to 80 pounds, representing a full process it used to be. Books aimed Journal of Aeronautical Sciences. Au-
scale span of 115 feet and a gross specifically at the homebuilder are now gust 1950.
weight of 40,000 pounds, respectively. available to ease the burden and im- 2. Schmitz, F. W. — "Aerodynamics
His model would be considered large prove the understanding. Two of which of the Model Airplane, Part 1, Airfoil
in comparison with today's recreational come immediately to mind are Craw- Measurements". Translation Branch,
models. But it apparently paid off for ford's "A Practical Guide to Airplane Redstone Scientific Information Center,
him. Performance and Design" and Stroj- Research and Development Directo-
nik's "Low Power Laminar Aircraft De- rate. U. S. Army Missile Command,
Rate of Climb sign". Both are frequently advertised in Redstone Arsenal, Alabama 35809.
Rate of climb is normally expressed Sport Aviation. 3. Hoerner, S. F. — "Fluid Dynamic
in feet per minute; a velocity. Thus, full The higher credibility of clues to iner- Drag". Published by the author. Avail-
scale climb rate would equate to the tial behavior comes in good measure able in most technical libraries. Also de-
model's climb rate times the square root from the fact that Reynolds Number is termine publisher's latest address
of X. seldom a factor of consequence. Thus, fromthe American Institute of Aeronau-
As shown in Table 1, if a 1/5 scale if the model's size, power, weight and tics and Astronautics (AIAA) in New
model were to show a climb rate of 344 moments of inertia are accurately York or Los Angeles.
SPORT AVIATION 35
 heed QT-2 Quiet Reconnaissance Air-
Table 3 • X Raised to Powers Used in This Article craft (which he accompanied to Viet
(For Users Having Calculators of Limited Capability) Nam as a company representative) . . .
was manager of airframe design and
experimental flight testing of the Lock-
X A X2 X3 X35 X4 X5 heed YO-3A (he holds the design pa-
4 2.00 16 64 128 256 1074
tent for the quiet recon aircraft). . . and
5 2.24 25 125 280 625 3125
6 2.45 36 216 529 1296 7776 was staff engineer on the Navy/Lock-
7 2.65 49 343 907 2401 16807 heed Surface Effect Ship program.
8 2.83 64 512 1448 4096 32768 Today, Stan is an aviation consultant,
9 3.00 81 729 2187 6561 59049 approved as an engineering subcon-
10 3.16 100 1000 3162 10000 100000 tractor to Lockheed Missiles and Space
Company, in which capacity, he has
done 38 conceptual designs for RPVs,
some of which were solar powered.
About the Author years, he has been a member of the Amazingly, in the midst of such a
Stan Hall is a long time member of engineering teams that designed such busy professional career, Stan has
EAA and has been a frequent con- notable aircraft as the North American somehow found the time to also fly as
tributor to Sport Aviation over the AT-6, B-25 and P-51; the Northrop XB- a corporate pilot and to personally de-
years . . . but that is only the tiniest tip 35 and YB-49 Flying Wings, the XP-61 sign, build and fly 10 aircraft of his own.
of the iceberg in his long and illustrious and XP-79 fighter bombers and the SX- He is an internationally known sailplane
career. A professional aircraft designer 4 research aircraft; and the Douglas designer (EAAers built scores of his
since 1940, Stan has specialized in pre- SCG-8 and SCG-15 cargo gliders. With Cherokees) and has been a member of
liminary and conceptual design, with the coming of the space age, Stan was the Soaring Society of America's Hall of
capabilities in structural and a member of the engineering team that Fame since 1974. Today, he is an ac-
aerodynamic analysis, actual construc- designed the Lockheed Agena space- tive, 5000 hour commercially certifi-
tion and piloting. Through employment craft . . . was project leader to the team cated pilot, with multi-engine, instru-
with several aerospace giants over the that designed, built and flew the Lock- ment and glider ratings.

£44 Membership Honor Roll

This month we continue our recognition of persons who have qualified for the EAA Membership Honor Roll. When you receive your new or
renewal EAA Membership Card, the reverse side of the attached form will contain an application with which you can sign up a new member.
Fill in your new member's name, enclose a check or money order and return to EAA Headquarters and you will be recognized on this page
in SPORT AVIATION — and there is no limit to how many times you may be so honored here.
Introduce your friends to the wonderful world of EAA . . . and be recognized for your effort. The following list contains names received through
the months of June 10.

JAMES B. ROSTER STANLEY PREDKO WILLIAM G. GORBY

Milwaukee, Wl Muskegon, Ml Hagerstown, MD
DENNIS G. WILLIAMSON ANTHONY J. CICERO MARK A. YOUNG
Madison, Wl Slatington, PA Lake Worth, FL
THOMAS H. IRLBECK ROSS SCHLABACH RAYMOND L. MUCHA
Somerset, Wl Taylors, SC Temple, TX
ROBERT A. KRUPKA WILLIAM E. GREEN MARC TILLIA
New York, NY Barren, Wl Bahia, Brazil
JOHN S. TUMILOWICZ ROGER CANNON JAY G. HALL
Wadding River, NY Wausau, Wl New Brighton, MN
CARL F. BACHLE LARRY A. BURTON JOSEPH M. LUCIA III
Jackson, Ml Klamath Falls, OR Seattle, WA
PATSY CUTRONE JOHN Y. BURCKHART C. DAVID SNARE
West Norwalk, CT Lehigh Acres, FL Shirleysburg, PA
THOMAS D. MILTON HENRY J. DEWALL JOSEPH K. LARRIMORE
Lansing, IL Escondido, CA Harrington, DE
PAUL HEDIGER MAURICE BOYER RANDAL B. GARDEN
St. Louis, MO Valleyfield, Que , Canada Lawton, OK
ARTHUR ROBERT BELOW, JR. VIC MARTIN MARTIN KAUFMANN
Wichita, KS Whitestone, NY Zurich, Switzerland
JOHN W. CLIFFORD E. JAMES HETTINGER BARBARA WRIGHT
Manassas, VA Janesville, Wl Newburyport, MA
J. JAY BILLMAYER DAVID M. BREGGER WILLIAM E. SCHADLER
Kalispell, MT Redlands, CA Fredericksburg, PA
ARTHUR S. RAYHLE RAYMOND B. SHERWOOD TUGDUAL BERTHO
Fort Mitchell, KY Pleasant Hill, CA Cachan, France
EDWARD CLEMENT ACRES RUFUS V. HOWARD JOHN R. SINGER
Adelaide, So. Australia Waukegan, IL Manitowoc, Wl
WILLIAM D. JOHNSON RALPH W. HAZELSWART
Hayward, CA Alexandria, VA
36 JULY 1987