Way back in 1975, my friend, Jerry Flynn, and I
(Kirt Blattenberger) assisted Dick Weber in his successful
flight on June 14, 1975, that set a new FAI Closed Course Record of 225 miles in 5 hours
and 38 minutes. We were both flaggers to signal when the Tortoise has passed the distance
markers. See the credits on page 37 in the actual magazine. The Tortoise got its name
by virtue of the craft having landed near a turtle on the runway. It was a long day,
with everyone being sunburned by the end of it. We were all members of the Prince Georges
Radio Control Club (PGRC).
Note: This article on the Tortoise
was previously posted with the magazine pages in image format. Now, this version has
been OCR'ed (optical character recognition) so that all of the text is searchable.
652 Miles Per Gallon!
Richard R. Weber
Setting a new FAI Closed Course Record of 225 miles in 5 hours and 38 minutes,
the Tortoise landed with 55% of its fuel still in the tanks.
In the autumn of 1973 I decided to go after FAI distance and duration records. I had
completed a satisfying summer of pylon racing, but there were not enough Quarter Midget
races in my area. World records offered a new challenge.

After setting the record, the designer displays the airplane and complete
support system. Size of the ship is only slightly larger than pattern jobs. Covering
is transparent Super MonoKote. Wing is half blue and half yellow, colors which show up
best against any sky background.
The first order of business was to find an engine that would run economically, so
engine tests were initiated. From the outset I intended to use a diesel engine because
it runs longer than glow, without the complication and interference of ignition. However,
most of the engines tested were glow, since they were available. An engine that runs
economically on glow fuel should be even better as a diesel. Articles were found on records
set by Bertrand, Giertz, Hill, Hirota, Kaiser, Reed and later Giertz. The engines they
used ranged from .15 to .49 cu. in. displacement. After many hours of running about 20
different engines, I chose a Supertigre 29 RV. One of the reasons for this choice was
that since the same basic crankcase is used up to a .46, it seemed that it should be
strong enough to convert to a diesel.
Next it was time to bone up on aerodynamics. I gathered many books and articles on
the subject. Perhaps the most useful book found was Man Powered Flight by Keith Sherwin.
The problems it considers are quite similar to those encountered in distance or duration
models, viz., designing to fly with minimum power. Another valuable book for the dedicated
modeler is Fluid-Dynamic Drag by S. F. Hoerner, which contains the results of a great
many airplane drag experiments relevant to model design. A third source of information
deserving more attention than we give it is the collection of NACA annual reports from
the 1920's and 1930's.
There are three record categories stressing fuel economy: Duration, Closed Course
Distance, and Straight Line Distance. Closed course distance was chosen first because
it requires a smaller plane than Duration and is a much easier undertaking than driving
several hundred miles (and back) for Straight Line Distance. The Straight Line Distance
record also requires more reliability, because the landing point must be specified within
500 meters before take-off.

Preparing for the flight the author is engrossed with the plumbing.
Tanks. two 48-oz. plastic detergent bottles extend 3 in. in front of wing to 5 in. behind.
Radio is midway between the wing and stab with hatch on top of fuse.
The airplane went through numerous design stages, keeping in mind the FAI limitations.
These include a maximum weight of 11.02 pounds, maximum flying surface loading (wing
and horizontal tail) of 24.57 ounces per square foot, and maximum engine displacement
of .61 in3. Early sketches were beautifully streamlined, but by the time construction
began the design had become rather simple, not unlike its record-setting predecessors.
Its chief features are the tapered wing with a genuine Clark Y airfoil, and a built-up
fuselage with no formers or cross braces.

A hefty shove is an important take-off assist. On record flight the
run was about 600 feet. Plane weighs 5 1/2 lbs. without fuel, 10 1/2 loaded.
The fuselage is filled from firewall to tail with fuel and radio components. There
are two fuel tanks, made from 48 oz. plastic detergent bottles, extending from 3" in
front of the wing to 5" behind. A hatch between the wing and stabilizer provides access
to the radio. The model is covered with transparent Super MonoKote. To assist visibility
under various sky conditions one wing panel is dark blue and the other is yellow. Experience
has shown that one color or the other is easily seen against any sky background. The
complete airplane weighs 5 1/2 pounds without fuel and nearly 10½ pounds fully fueled.

Radio was Kraft 6-channel with two-third brick in airplane. Engine
was Super Tigre R/C .15 Diesel conversion. Glow case proved inadequate for sustained
Diesel running.
The radio control system had to be reliable, compact, and easy on the battery. A check
of the current drain of several systems indicated that the Kraft 3-channel brick would
be ideal. I obtained a brick which was then converted to my transmitter frequency of
53.2 MHz. Some additional work by Doug Spreng of Kraft Systems reduced the servo current
drain further. I planned to use rudder and elevator controls, but the third channel would
be available for engine control if needed.

Close up of the special field box containing the interlace equipment
described in article. A compact arrangement.
The batteries for the airborne radio system were four C cell NiCads, rated at 1.5
AH and found to exceed 2 AH. The NiCad transmitter battery pack is good for only four
or five hours, so I made an interface box to charge the transmitter batteries and power
the transmitter simultaneously. It contains meters to monitor the current and voltage
supplied to the transmitter. It also has a potentiometer to vary the current. This interface
box connects the transmitter to the battery in my field box or to a car battery. It was
used about half the time during the record flight.

During the summer of 1974 I flew RC planes with diesel engines in order to become
familiar with their characteristics. I would like to encourage much wider use of diesels.
Contrary to popular belief, diesels start easily and they can be throttled down like
the best glow engines. Perhaps my most notable flight of 1974 was on September 2 with
a heavy Box Fly Jr. powered by a Supertigre 15 R/C diesel. It took off the ground and
flew with the engine running for 1 hour and 45 minutes on four ounces of fuel.

Result of ground loop ending in grass was snapped fuselage. Repairs
took only a half hour. Controls were elevator. rudder and needle valve: Pressure fuel
system was used with homemade regulator between tanks and needle valve. Four C cell NiCads
supplied airborne radio. Transmitter pack good for only four or five hours. Interface
box permitted simultaneous charging of batteries, power to X-mitter.
Initial flight tests on the new record model began on December 23, 1974. A fuel cutoff
was connected to the third servo for ending test hops. I tried a variety of likely propellers
with the plane carrying two pounds of water and 11 ounces of fuel. Surprisingly little
difference in air speed was found between the different test props. The final choice
was a Power Prop 11-7½, just like on your big pattern ship. Later flights gradually increased
the fuel load, to ascertain whether the shifting CG and increased weight would present
any problems; generally they did not.
The fuel system consisted of two pressurized plastic tanks connected in series, followed
by a fuel pressure regulator. Tests were all satisfactory, although a problem developed
later.

On May 10, 1975, we gathered for an official attempt on the FAI Closed Course Distance
record, sponsored by the Prince Georges RC Club and the Goddard MAC.
Present were John Sites, CD, Luther Jackson, Eric Baugher, Ken Greenhouse, Ron Moltz
and Chet Opal. I estimated the chance of success at 50-50, since we knew of no problems
but the plane had never been flown longer than 75 minutes. It was not a propitious day.
At take-off the 10- plus-pound airplane ground looped and went into the grass beside
the runway. The fuselage broke in half behind the wing. A brief discussion ensued and
Ron Moltz was off to his home for some Hot Stuff and sticky MonoKote. To my amazement
the plane was repaired and flying half an hour after he returned. The official flight
was only fifteen minutes old when the engine went lean and slowed down. Then it repeatedly
improved and went lean in random fashion. We worried about it when it was very lean,
but it kept on running. Lap times, which had been running 52-54 seconds in the beginning,
soared to over 70 seconds, but still it ran. After an hour of this I had just decided
to stop worrying, convinced that it would continue to torment us all day long, when the
engine stopped suddenly after 73 laps.

Tortoise airframe details.
When the plane landed we found that the rear engine bearing had pushed out the bottom
of the crankcase. We theorized about several possible causes of the problem: faulty regulator,
lean needle-valve setting causing overheating and deposits on the piston crown, which
raised the compression too high, or fatigue from detonation during earlier bench testing
of the engine. The possible problems were all rectified. The regulator was reworked,
the needle valve was hooked up to servo control and a brand new crankcase was found.
The second attempt on the record was on June 14, 1975. The officials were Luther Jackson,
CD, Dean Smith, Chet Opal, Eric Baugher, John Tallman, Jerry Flynn, and
Kirt Blattenberger.
After one of the test flights to adjust the engine, the plane landed near a tortoise
walking on the runway. We immediately concluded that this was a good omen. Since the
plane was nameless, it was dubbed the Tortoise, for its slow but steady flight.
When the tanks were filled the ground looping of May reappeared, but the plane took
off on the fourth attempt, after a ground run of about 600 feet. The old record was 338
laps (kilometers) around the course defined by two pylons 500 meters apart. Since FAI
records must be broken by 2%, our magic number was 345 laps. The flight went without
major problems. Occasionally the motor would go lean, but the new mixture control took
care of this. Sometimes when the motor sounded fine, I would set it leaner to see if
it would sag; usually it did and was immediately set richer again. As we were nearing
the record, at about lap 341, the engine went quite lean, but the needle valve was quickly
opened and each of us held his breath. When lap 345 was completed, there was a round
of cheers, for the record was ours!
The engine then seemed happier, probably because we were not so concerned. Talk of
400 and 500 laps began to sound reasonable, but it ended abruptly at the end of lap 363,
when the engine stopped without warning. The plane glided in to a smooth landing at the
center of the course. It had flown 363 kilometers (225 miles) in 5 hours and 38 minutes,
averaging 40 mph.
Upon removing the wing we were overjoyed to see what looked like 75% of the fuel remaining!
There would be another day and a longer record for the Tortoise! Then we noticed that
a check valve in the pressure system had failed to allow air into the tanks to replace
the fuel removed, and the tanks were partially collapsed from engine suction. A subsequent
measurement showed that 55% of the fuel actually remained. This works out to about 650
miles per gallon.
The problem which stopped the plane was found to be the same one we had in May: the
bottom of the crankcase broke under the main bearing. The reason for the problem is now
clear. The high-compression loads of diesel operation exceed those which the crankcase
can tolerate for extended periods of operation. If a stronger engine can be found, and
I believe it can, the plane should be capable of 500 miles on a closed course and more
on a straight line course. With a larger wing and lower rpms, maybe it can exceed the
present duration record of 14 1/2 hours set last year by Lars Giertz of Houston.
An assault on world records cannot be made without the help of many people. My greatest
help came from Dean Smith, whose engineering expertise and machining abilities were indispensible.
Dean and I kicked around many ideas and went down our share of dead ends before finding
the right combination. Dean made the pressure regulator, the parts for the diesel conversions,
and many other parts used in the assorted experiments we tried. I also benefitted from
discussions with Don Jehlik and Cliff Telford, both world champions and gentlemen.
Editor's Note: A straight-line distance RC performance record of approximately 266
miles was established by the author on August 16. The record is in the process of being
homologated by FAI. Take-off was at 9:31 a.m. from Newton, Kan. and the landing took
place at 4:49 p.m., at Enole, Nebr.
Posted March 23, 2012
|