|
|
Trainer Design
By reducing the dimensions of a full-sized aircraft proportionally, a scaled
model will be obtained, however, it seldom becomes an easy flying one.
The main aerodynamic differences between a model and a full-sized aircraft
are originated from the boundary layer, the thin layer of air close to the wing
surface that is slowed down by skin friction.
According to Osborne Reynolds, there are two main types of flow:
The laminar and the turbulent.
Which flow type occurs within the boundary layer at a given point of the wing's
surface depends on the wing's form, the surface's roughness, the chord lenght,
the airspeed and the ratio of density to viscosity of the air.
Reynolds combined all those factors (except the surface condition) into a non-
dimensional number known as Reynolds Number Re.
Re = (air density/air viscosity) * air speed * wing chord
Air viscosity is measured in kilograms per meter per second.
The standard value is: 0.0000179 kg/m/sec.
For instance, a wing with a chord of 1 meter at a airspeed of 1 m/sec
and with the standard air density and viscosity will have the following Re:
(1.225/0.0000179) * 1 * 1 = 68459
Thereby, a simplified formula may be obtained as follows:
Re = 68459 * V * L
Where V is the airspeed in m/sec and L the wing chord in meters.
The Reynolds number is therefore dependent on the weather conditions, the
wing chord and the airspeed.
Re increases as the airspeed, the air density and the wing chord increases.
Since the wing chords of model aircraft are often much less than 1 meter, one
may get a Re value close enough for modelling purposes by using the following
simplified formula:
Re = speed in kilometres per hour * chord in centimetres * 189 (Metric units).
Re = speed in miles per hour * chord in inches * 770 (Imperial units).
At low airspeed and small wing chord (as with a model aircraft) the air viscosity
is a dominant factor, whereas with the full-sized aircraft the viscosity effects of
the air are insignificant while the aircraft's mass inertia becomes more dominant.
That's why one should not expect a scaled model aircraft to have the same flight
characteristics as its larger counterpart.
As stated in Forces in Flight, the lift force is dependent on the density of the air r,
the airspeed V, the wing's Lift Coefficient and the wing抯 area according to the
formula:
Lift Force = 0.5 * r * V2 * Wing's Lift Coefficient * Wing's Area
The Wing's Lift Coefficient is a dimensionless number that depends on the airfoil
type, the wing's aspect ratio (AR), Reynolds Number (Re) and is proportional to
the angle of attack (AoA) before reaching the stall angle.
However, the wing's generation of lift also produces Induced Drag, which together
with Parasitic Drag are forces that oppose the aircraft's motion through the air.
So, one may say that Induced Drag is the price we pay for getting lift.
Induced Drag is also dependent on the density of the air r, the airspeed V, the wing's
Drag Coefficient and the wing抯 area according to the formula:
Drag Force = 0.5 * r * V2 * Wing's Drag Coefficient * Wing's Area
The Wing's Drag Coefficient is a dimensionless number that depends on the airfoil
type, the wing's aspect ratio (AR), the shape of the wing tips, Reynolds Number (Re)
and the angle of attack (AoA).
The relation between lift and drag is called the Lift to Drag ratio (L/D) and is obtained
by dividing the Lift Coefficient by the Drag Coefficient.
The characteristics of any particular airfoil may be represented by graphs showing the
amount of lift and drag obtained at various angles of attack as well as the Lift/Drag ratio.
The same airfoil has different Lift Coefficients and different Lift/Drag ratios at different
Reynolds Numbers as shown in the graphs below:
The best Lift/Drag ratio is where the green lines touch the curves in the graph on right.
The above graphs refer to the airfoil only, as the Lift Coefficient of a whole wing also depends on
the wing's Aspect Ratio and the shape of the wing tips.
 |
The graph on left shows Lift and Drag
Coefficients along with Lift/Drag ratio
of a whole wing with aspect ratio of 9
and airfoil RAF 32 at Re 56,100.
The Max Lift Coefficient is obtained at
about 9.2o AoA, while the best L/D is
obtained at 3o AoA. |
A large wing that is flying fast has a higher Re and thinner boundary layer than a small
wing that is flying slow. The boundary layer is thinnest when its flow is laminar and
thickens when it is turbulent.
The turbulent flow may separate from the wing's surface, producing more drag and
decreasing the lift, which may lead to stall.
Thus, a low Re wing is more likely to suffer from laminar separation and to stall sooner
than a wing with high Re.
| Typical Reynolds Numbers:[/td] | | Full scale airliner | above 10 000 000 | | Light aircraft | above 1 000 000 | | Large model aircraft | less than 400 000 | | Typical model aircraft | less than 200 000 | | Indoors and slow flyers | less than 30 000 |
The area of the flying surfaces (wings, fin and stabiliser) as well as the control
surfaces (elevator, rudder and ailerons) should be proportionally larger in the
model aircraft in order to obtain more controllable flights and landings.
Wing loading is also more critical with smaller models. That means, a bigger
model may have greater wing loading than a smaller one.
Some basic rules of thumb may be followed when designing an easy flying
trainer model according to the pictures below:
One may start by choosing the desired wing chord or wingspan from which all
other related dimensions may be calculated.
With high wings the dihedral angle is typically between 3 to 6 degrees.
Dihedral should be lower when using ailerons (up to 3 deg).
Although not strictly needed, a washout angle between 3 to 5 degrees is
advisable in order to improve stall characteristics.
The ideal incidence and motor thrust angles are usually found by trial and error.
Initially, one may start with 2 to 3 degrees down and right thrust.
The wing's and stabiliser's incidence may preliminary be set at zero, and may
be changed during test flights.
Flat bottom wings may need more down thrust than symmetrical and/or semi-
symmetrical ones.
Landing gear placement on a tail dragger should have the axle coincident with
the leading edge of the wing, whereas on a tricycle the main gear should be
slightly aft of the CG balance point in order to get easier take-offs.
A tail-heavy aircraft will be more unstable and susceptible to stall at low speed
e. g. during the landing approach.
A nose-heavy aircraft will be more difficult to takeoff from the ground and to
gain altitude and will tend to drop its nose when the throttle is reduced. It also
requires higher speed in order to land safely.
Recommended Engine Size vs
Wing Area | | c. c. | c. in. | area sq. dm | area sq. in. | | 0.8 | .049 | 12 - 16 | 200 - 250 | | 1.6 | .10 | 15 - 22 | 250 - 350 | | 2.5 | .15 | 20 - 30 | 300 - 450 | | 4.0 | .25 | 26 - 32 | 400 - 500 | | 6.7 | .40 | 32 - 45 | 500 - 700 | | 10 | .60 | 38 - 55 | 600 - 850 |
Powered model aircraft performance may also be estimated by calculating the
weight / power ratio, also known as power loading.
A slow and low wing loading (for a beginner), with a weight / power ratio of 440
to 500g/c.c. (270 to 300oz/c.in.) might be good enough, whereas an aerobatics
would need about 340g/c.c. (200oz/c.in.) to achieve good performance.
This is assuming 2-stroke engines and that the power of different types is pro-
portional to their displacements, (which isn't too far off).
As for the airfoil type, one should consider that a flat bottom wing gives high lift at
upright flight but poor lift at inverted flight.
Flat bottom wings (high cambered airfoils) are mainly used in slow and relatively
light powered models. They have high lift coefficient but also high pitching moment,
so a relatively longer tail moment or larger stab area may be needed in order to
achieve a good longitudinal stability (stability in pitch).
They also tend to balloon when power is increased or when turning into the wind.
Quasi-symmetrical airfoils are usually a good compromise giving almost the same
lift at both upright and inverted flight.
Symmetrical airfoils are intended for aerobatic models as it behaves equal at both
upright and inverted flight.
The control surfaces' max throws also have great effect on the flight stability.
With a too much throw the model will respond too quickly and may be difficult
to control, whereas too little throw will result in poor control, especially at low
landing speed.
Typical throw settings measured at the control surface trailing edge are:
Elevator and Ailerons 6mm (1/4") up and down.
Differential Ailerons (recommended with flat bottom wings) 8mm (5/16") up
and 4mm (5/32") down.
Rudder 10mm (3/8") left and right.
Those figures are just guidelines and some minor changes may be done
during test flights. Faster models will require lower throw settings.
To increase the control surface throw, move the push rod to the hole on the
control horn that is closer to the control surface and/or move the push rod to the
further out hole on the servo arm.
Some transmitters have dual rate facility, which allows the pilot to change the
max throws to suit the flying speed.
The picture below shows a typical radio installation. Both the battery and the
receiver are wrapped upp in soft foam to damp the motor vibrations.
The material and construction methods depend on the model itself, personal
preferences and on materials and tools available.
Given enough power, almost anything can be made to fly - the question is how...
One should also bear in mind that landing is an inevitable part of the flight.
In order to avoid stall, a plane with high wing loading requires higher take-off and
landing speed.
Two planes with different sizes and with the same wing loading will have about
the same stall speed, but the smaller one will seem to fly faster and will be more
difficult to control, especially during landing approach.
Thus, one should strive to build as light and as strong as possible.
Typical wing loading with a 150cm wingspan (60 in) model is about 60g/sq.dm
(19-oz/sq. ft). This value may be slightly higher with bigger models but should
definitely be lower with smaller ones.
For instance, the wing loading of a full-scale Cessna 152 is about 510g/sq.dm
(167-oz/sq.ft), a model aircraft with such a wing loading would hardly be able to fly.
Wing loading is the aircraft's weight divided by the wing area.
| Calculate Wing Loading & Area | | Wingspan: | inches
| mm
| | Wing Root Chord: | inches
| mm
| | Wing Tip Chord: | inches
| mm
| | Or Average Chord: | inches
| mm
| | Or Wing Area: | sq.in
| sq.dm
| | Model Weight: | ounces
| grams
| | | | WING LOADING: | oz/sq.ft
| g/sq.dm
| [/tr]| CUBIC LOADING: | oz/cu.ft.
| | |
Elliptical wings' area can be calculated as follows:
A reference that is not dependent on the aircraft size is the cubic wing loading,
which is calculated dividing the weight by the wing area raised to the 1.5 power.
For instance, the full scale Cessna has a cubic loading of about 13 oz/cu.ft, which
puts it at the high end of a scale model category regardless of size.
Different types of model aircraft may have different cubic wing loadings (oz/cu.ft)
as shown below:
| Model Type | Cubic Loading | | Sail and Park Flyer: | 4 to 7 | | Sport and Trainer: | 7 to 9 | | Pylon and Scale: | up to 13 | | Electric Ducted Fan: | up to 25 |
Beginners are advised to choose cubic wing loading values no greater than 8, as
it's likely to give relatively low take-off and landing speeds.
At higher cubic loadings one should expect increased landing and take-off speeds
assuming no special lift devices are used, such as flaps.
As a rule of thumb the stall speed in mph is approximately equal to four times the
square root of the wingloading in ounces per square foot.
But if you know the whole wing's max lift coefficient, you may get a more accurate
result with following formula:
Stall speed (m/s) = [2*Weight / (Clmax*1.225*Wing Area)] 0.5
Where the weight is in Newton, area in m2 and standard air density 1.225kg/m3
Unless it's a glider, the prop's static pitch speed should be higher than 2.5 times
the aircraft's stall speed.
The static thrust should be at least about 1/3 of the planes' weight in order to get
reasonable climb and acceleration capabilities after aborted landings.
Some scaling rules:
A scale model's weight should be reduced by the cube of the scale factor.
For instance, a full-scale Piper J-3 Cub weights 1000lb and has 36ft wingspan.
A 1/6 scale Piper J-3 Cub model should weight 1000/63 = 4.6lb.
The wingloading of the scale model should be reduced by the scale factor's ratio.
The 1/6 scale model should have 13.3 oz/sq.ft wingloading instead of 80 oz/sq.ft
as the full-scale Piper J-3 Cub.
Also to get a "scale-like" visual appearance in flight of scale models, one might
reduce the speed of the full-scale aircraft by the scale factor's ratio in order to get
a linearly scaled speed.
For example, the full-scale J-3 Cub cruises at about 70 mph, thus a 1/2 size Cub
should cruise at about 35 mph. Both the model and the full-scale aircraft should
move the same number of fuselage-lengths per second, and would appear to the
eyes of the observer to be flying at the same speed.
However, this is not so easy to achieve, especially with smaller models outdoors.
At 1/2 of the speed, the wing's lift (even ignoring Re effect) is only 1/16 as much.
But, since the air molecules don't scale down when we reduce the plane's size,
flying half as fast with half the size results in 1/4 Re, which further reduces the lift.
Also the wind is not scaled down either, that means a 20 mph gust for a model is
much more serious than 20 mph gust for the full-scale.
So, flying at scale speed isn't so practical for small models, unless flying indoors.
Some unit conversions:
1ft = 0.3048m
1in = 2.54cm
1lb = 16oz = 0.4536kg
1oz = 28.35g
1sq. ft = 144 sq. in
Multiplying lb/sq.in by 2304 gives the value in oz/sq. ft.
To download a complete Unit's Converter click on the picture below:
Some interesting books:
- for further details click on the pictures -
   
   
|
|
/1 
发表于 2009-4-23 11:15
收藏
发表于 2009-4-23 12:45
楼主