|
Winning the National Radio Control Meet
The take-off! Practically all flights are made in this manner
to ensure good take-offs despite the man at the controls. The plane
is being guided, not lifted .
Details of a Radio-Controlled Model Airplane That Has Made Over
One Hundred Successful Flights
By WM. E. Good, W8IFD
Four years ago a radio-control event was added to the program
of the annual National Model Aircraft Championship Meet. But it
was not until last year that a wholly successful demonstration of
radio-controlled flight under a variety of conditions was finally
achieved. The Good brothers of Kalamazoo were responsible, their
triumph climaxing years of experimentation. Here is the story of
their success.
One of the most interesting by-products of ham radio is that
of radio control of gasoline-powered model airplanes. Here is one
sport where your signals must be QSA 5 or you may have to start
on a cross-country jaunt, praying that those two ounces of gasoline
will hurry up and run out. If your signals are "getting through,"
you find yourself landing the radio-controlled plane right in the
middle of the runway. Exciting? I'll say! Every flight is just like
that memorable first QSO!
Radio control of model planes began with the advent of small
gasoline motors and has been progressing slowly ever since. Most
of the development has been done by hams in cooperation with the
gas model enthusiasts.1
Our equipment is the result of over four years' experimentation
and of late has proved rather successful. The control consists of
two frequency channels - one for the rudder and the other for the
elevator. For each channel there is a modified five-meter superregenerative
receiver. In its plate circuit is a sensitive relay which is connected
to an electromagnetically-operated rubber-powered escapement in
the tail which moves the controlling surface in the fashion desired.
The plane, designed and built by my twin brother Walter, has
an 8-foot wing span and weighs slightly over eight pounds, including
its two pounds of radio gear. Its gasoline motor is of the one-fifth
horsepower variety and does a good job of flying this stable ship.
The cruising speed is about 20 miles per hour and the ship glides
very well, insuring excellent landings if reasonable skill is used
at the controls. The plane banks automatically when the rudder is
turned, due to the dihedral in the main wing.
Fig. 1 - Radio Control Receiver.
T - RK42 (not RK-62).
C1 - 250-μμfd. midget mica.
C2 - 15-μμfd. midget variable.
C3 - 0.002-μfd. midget mica.
C4 - 0.005-μfd. midget mica.
C5 - 30-μμfd. mica trimmer.
L1 - Each half 5 turns No. 14 or smaller wound on
3/8" diameter form, spaced wire diameter.
L2 - Interruption-frequency oscillator coil (National
OSR).
B1 - 1.5-volt flashlight cell.
B2 - 45-volt midget 10-oz• "B" battery.
R1 -10,000-ohm midget variable.
Relay - See text and Fig. 3.
Before a flight the receiver in each channel in the plane is
adjusted so that its sensitive relay closes when the carrier from
the five-meter transmitter is turned on. The sensitive-relay contacts
actuate the small electromagnet (in the tail) which allows the rubber-powered
escapement wheel to go through one position or one-quarter of a
revolution at a time (i.e., for each dash sent). The controlling
surface is connected directly by a small steel-wire arm to a pin
on the escapement wheel. The power used to move the surface through
its positions is taken from the wound-up rubber band. Our control
surfaces have three main positions, e.g., the rudder has left, neutral
and right, plus two half or intermediate positions, making five
in all. Naturally, the movements take place in a cyclic fashion.
Each pulse or dash from the transmitter causes the surface to move
from one main position to the next.
Walter has done practically all of the design and detail work
on the escapements and the sensitive relays, although he's not even
a ham. We did have him in the RI's office one Saturday morning to
take his exam - but he sneaked out and made the rounds of the model
airplane shops in Chicago!
In flight! Framed by the center-fed doublet atop a bamboo mast,
the ship is but a speck in the sky. The radio controls operate as
far as the ship can be seen.
Escapements Located in Tail
The escapement units in their present state weigh just a half-ounce
each and are mounted permanently in the tail surfaces, so that direct
mechanical connection can be made to the moving elements (rudder
and elevator). This makes for extreme reliability in addition to
the fact that the units boast almost instantaneous response, which
has been shown to be practically a necessity under actual flying
conditions. A control stick or "joy stick" (one for each control)
which adapts itself very well to the escapements is a Western Electric
telephone switch which was rebuilt so that contact is made and broken
(sending a dash) as the switch is moved from neutral to either extreme
position. Thus the rudder or elevator will be in the same position
as its corresponding control "stick," and this synchronization will
be maintained as long as the control switches are moved through
complete cycles. The moving elements will follow the motions of
the switches even though they are jerked back and forth as fast
as four or five times a second. Thus the surface may be moved to
any desired position with such rapidity that the motion of the plane
is not affected while so doing. Due to the arrangement of the switches
and the tail escapements, the corresponding moving element will
be in a half or intermediate position when its control switch is
in a half position. This system allows the operator to know exactly
where the rudder and elevator are positioned at any instant. Actual
flying has shown this to be a natural method of control.
Top - Bill Good, W8IFD, with the fuselage and ground control
set-up. Left, the genemotor and its battery box; center, the two-frequency
transmitter in its travelling case; lower right, the control box
with two telephone-type switches as joy sticks.
Bottom - Ready to fly! Walter Good, national radio-control champion,
holding his winning ship. The receiver is accessible through the
open doorway in the side of the cabin.
The sensitive relays have been the result of a generous mixture
of theory and experiment on the ground and in the air. In all, six
relays have been developed. The one that has proved the most satisfactory
and the one that we have been using for the past two years we call
the DG-6. (It can be duplicated for fifty cents.) It is a polarized
balanced-armature type (no springs), weighing two ounces, and operates
on less than one milliwatt. For vibrating motors and airplanes in
flight we believe this is the Good answer to the maiden's prayer!
The receivers have also had their share of attention, and considerable
research has been done to find the best operating conditions. Essentially
they are one-tube superregenerative jobs in the self-quenching circuit
using quench coils, but the conditions of operation have been changed
so that a maximum change in plate current takes place when a signal
is received. A Type 30 tube will give a plate current change of
as much as three milliamperes, while an RK42 (a 1.5-volt version
of the 30) will give about two milliamperes change (through 2500
ohms). Besides the tuning condenser, the antenna padder and the
grid bias resistor are adjusted to obtain maximum plate current
change. On the field it is only occasionally necessary to adjust
the variable grid resistor. Experimenting has been done with various
circuit constants and different makes of quench coils in addition
to different quench frequencies and quench voltages. By extending
the recently developed superregeneration theory, it is not difficult
to explain the theoretical basis for the required adjustments.
C1, C2, C7 - 100-μμfd. midget
mica.
C3 - 35-μμfd. midget variable.
C4, C5, C6, C8, C9
- 0.002-μfd. midget mica.
C10 - 15-μfd. midget variable.
R1 - 0.1-megohm, 31-watt.
R2 - 50,000 ohms, 2-watt.
R3 - 150,000 ohms, 1 watt.
R4 - 25,000 ohms, 1 watt.
R5 - 25,000 ohms, 10 watts.
-C - 45-volt midget "B" battery.
M1 - 100 ma.
M2 - 0.5ma.
L1 - 9 turns No. 14, tapped turn from ground.
L2 - 8 turns No. 14.
L3 - 6 turns No. 12.
All coils 3/4" dia., windings spaced wire diameter. L2
is tuned by squeezing or pulling out the winding.
The receiver in its balsa-wood container, with the battery supply
underneath. The miniature polarized relay can be seen in detail.
Batteries are attached rigidly to the fuselage, hut the receiver
is mounted in sponge rubber and suspended by rubber bands.
The Two-Frequency Transmitter
The transmitter is unique in several ways besides portability
- if you call lugging a six-volt storage battery portability! It
utilizes two separate electron-coupled oscillators (one frequency
for each channel) and only one final amplifier. The two 6V6G e.c.o.'s
have a common plate tank - which, by the way, works very well -
broadly tuned and capacity coupled to the 807 doubler-final. The
half-wave horizontal center fed Hertz antenna is furnished r.f.
by a tuned line which is inductively coupled to the final. A switch
is incorporated to remove the power from the 807, leaving just the
weak five-meter harmonics from the e.c.o.'s for tuning the receivers
when the plane is near the transmitter. The genemotor mounted on
top of the storage battery is a 400-volt 125-ma. job. Nevertheless,
only 20 to 30 watts of the fifty available is the accustomed power
input.
Fig. 2 - The Two-Frequency R/C Transmitter.
Fig. 3 - The DG-6 Sensitive Relay.
This polarized relay, designed and built by Walter Good, is of
the balanced armature type, having no springs on the armature. A
small Alnico horseshoe magnet holds the thin iron armature against
the back contact until the coil and its surrounding soft-iron magnetic
circuit is energized. With a 2500-ohm coil the reliable sensitivity
is 1 milliwatt. The weight is slightly over 2 oz.
When the plane is properly adjusted it flies and glides straight
with the neutral rudder position and gives the same size circles
to the right or left for the extreme positions. This is true both
under power and in the glide because of proper power and wing loading.
A great deal of flying is done with the rudder alone. For this,
the elevator is adjusted for a good climb and then it maintains
level flight in the turns. Theoretically, the gyroscopic effect
of the motor should cause the nose of the plane to come up when
the plane is turned left and down when turned right, but practically
this effect is not noticeable.
Fig. 4 - Isometric View of the Tail Assembly in the Good Championship
Plane.
The small control escapement is considerably exaggerated in size
to show detail. Actually the small magnet coils are about 1/4" diameter
and each escapement weighs about 1/2 oz. The elevator control escapement
is similar, being located in the stabilizer just to the right of
the fin juncture.
When the rudder is turned to either right or left maximum position,
the plane banks automatically and proceeds to execute a beautiful
right or left circle. If the control is kept in this position for
more than one turn or so, the bank gradually becomes steeper and
the turn develops into a large spiral. It is not difficult to lose
any amount of altitude in short order by spiraling the ship down
in this manner, even if it is only for the gratifying sensation
of pushing the control stick neutral to watch the plane straighten
out and start into a fast climb, using up the speed just acquired
by the descent.
One question that always arises in radio control discussions
is, "Is it necessary to have such speedy snap-action controls?"
In our flying we have found that fast control has been more than
convenient, especially in take-offs and landings. Many times in
coming in for a landing it has been imperative to give opposite
rudder to straighten out the glide when the ship was only four or
five feet from the ground. A second of time in such a predicament
is precious. Take-offs call for more precision and speed of control,
because the controls are more sensitive and the operator really
has to have the "feel" of the controls to keep the plane right-side
up. The picture shows the usual take-off procedure of running the
wing tips until the plane is well off the ground. We've found it's
much safer to learn how to "fly" after the plane is up in the air!
Lately, however, we have been merely starting the plane down the
runway (tail off, but wheels still touching) and keeping it as straight
as possible with the rudder control. Strangely enough, when the
plane starts towards the edge of the runway (and it usually does)
plenty of control is needed instantly to bring it back and then
care must be taken not to over-control. Thrilling? Yes! But if you
control wrongly a wing tip starts digging up the runway or vice
versa!
The National Championship Meet
In winning the Radio Control event at the National Model Airplane
Contest at Detroit this year, the radio and the plane performed
in grand style. The radio-control planes were judged on their ability
to execute a number of pre-decided maneuvers. The best flight we
had lasted about 14 minutes. The ship climbed to approximately 1500
feet during the six minutes the motor was running. During the first
part of the flight the model was sent down-wind to a field-light
objective about one-quarter of a mile away, following the judge's
instructions, and then was turned around and brought back over the
transmitter. As you can see, this is an excellent stunt really to
test the controllability of the model. Next, as we usually do, the
plane was guided up-wind and the rest of the flight consisted of
right and left circles, figure eights and the like, on command of
the judge. At the end of this particular flight the job was landed
about a hundred feet from the transmitter, thus establishing the
first real radio-controlled flight at a National Contest.
Will any transmitter work the control? Yes, anyone that's on
the right frequency in the five-meter band. The five-meter gang
from Detroit was on hand at the Nationals to help with the radio-control
event and their main transmitter - a portable-mobile outfit - operated
our controls very effectively. However, the cooperation between
the contestants and operators was very gratifying and for the most
part no interference resulted.
The only case of interference we've had in over a hundred flights
this year and about fifty flights last year occurred in Chicago
during a demonstration at the big Mid-Western States gas-model contest.
We had flown the job in a strong wind in the morning and Walter
had succeeded in landing the plane within two feet of the point
where the wheels had left the ground on the take-off! Naturally,
feeling so confident about the success in such windy weather, we
decided to send her up again in the afternoon. It was my turn to
"fly." Everything went fine until about thirty seconds after the
motor cut, when the plane refused to respond any more - 1200 feet
up, slightly upwind and a 15-mile "breeze" blowing! Nothing we did
on the ground had any effect - the plane was making great progress
cross-country in a large circle, indicating half rudder position.
Possibly another five-meter carrier was holding the relay down?
The plane eventually landed about a mile and a half away and was
finally recovered - but that's another story. The controls were
checked and found to be in working order, leaving us with only one
conclusion - that some amateur in the Chicago area was operating
on the same frequency! (Time of flight, 4 P.M., Aug. 6, 1939.)
This system of control has worked as long as the plane has been
in sight - about two miles. So far we've found no reason for flying
it at a greater distance than that, especially when our original
purpose was to bring the model back to the field so we wouldn't
have to chase it!
This plane has done itself proud by winning the "Nationals" two
years in a row, by taking the radio-control event in Chicago, by
receiving first place in the original-design event at the Scripps-Howard
Junior Air Races at Akron, by being possibly the first radio-controlled
plane to be flown in Canada through a number of flights made at
the Canadian National Contest at Toronto, and finally by its good
behavior during demonstration flights at contests around Michigan.
We hope we've worked up your enthusiasm so that you may join
this exciting diversion of amateur radio. All bragging aside, it's
not easy, but, boy - it's lots of fun!
1 QST, Oct. 1937 and June, Sept. and Oct. 1938; Model Airplane
News, January and August, 1938; Air Trails, August 1938, January
and May, 1939.
Posted
|
