From what I can remember, this October
1972 edition of American Aircraft Modeler is the first I received after joining the Academy
of Model Aeronautics (AMA). I was thrilled to be having a monthly modeling magazine delivered
to my rural home because it was rare that a copy of Flying Models or Model Airplane News
would appear on the rack in our local convenience store. Unlike today's age of instant
and ubiquitous information, getting ahold of desired reading material was not nearly
as easy before the Internet. Somehow, I managed to retain possession of that issue for
nearly 40 years now. With few exceptions, everything else from my childhood has vanished.
I remember being particularly interested in the Charybdis because it satisfied the
desire for a lot of different modeling interests - helicopters, airplanes, and nitro-powered
engines. In 1972 I was 14 years old and didn't have a lot of walking around money - only
what I scraped as profit from my paper delivery route and odd lawn mowing jobs. So, I
never actually built one, but I sure did pour over the plans and article intently.
Having forgotten about the Charybdis until running across it again, I'm thinking maybe
a modern, electric-powered version would make for a great winter project. Check back
here in the spring to see if I make good on my idea.
Charybdis
With a twist of your wrist, this single-bladed, .010-powered thing swishes aloft. Great
fun and easy to make.By H.D.M. Sherrerd, Jr.
In Greek mythology, Charybdis was an extremely powerful
whirlpool off the Sicilian coast. This helicopter is not particularly powerful, but like
its namesake, everything revolves at a rather high rate of speed, and to that degree
at least the name was appropriately chosen by the inventor, Charles W. McCutchen of Princeton,
New Jersey.
Charybdis was developed 18 years ago, while McCutchen was living in Cambridge, England,
and caused something of a sensation when he took it to the British Nationals of 1954.
Since that time, variations on the "McCutchen Machine," as the design is more generally
called, have occasionally appeared in European magazines, but a prophet is usually without
honor in his own country, and the Charybdis seems to have been completely ignored in
the U.S.-a pity, since it is no tougher to build than a good hand-launched glider, and
more fun than tying firecrackers to your old flying scales.
Construction
Construction is quite simple, with the emphasis on strength. The blade and stabilizer
are of sheet rather than built-up, the motor and balance arms are of spruce, and the
hub is reinforced with 1/16th plywood. The motor arm is inlaid into the lower surface
of the blade, the reinforcing plate double-glued over the joint, and the whole bound
with silk or other light cloth. This area is then virtually unbreakable, and also a good
flat surface for the balance arm to bear against. You can, of course, glue or even bolt
the balance arm in place, but Charybdis is much easier to carry around if the arm is
detachable.
The blade is simply a 2' x 2" lath of 1/4-in. medium sheet balsa, shaped to a constant
Clark Y section. No wash-in, no wash-out, no dihedral breaks; the squarest, easiest wing
you ever made. Use a template to maintain section accuracy. Don't try something with
undercamber instead. McCutchen tried both undercambered and curved-sheet airfoils, and
found the resulting Charybdis to be unimproved, at best, or just plain unstable, at worst.
The stabilizer struts are 1/8th hard sheet sanded to a streamline section. Use plenty
of glue and perhaps even some silk reinforcing at the strut-to-stabilizer joint, the
only vulnerable area of Charybdis. The stabilizer itself is also 1/8th hard sheet, but
is given a lifting section. Be careful to set it at an angle of at least -50 or -60 relative
to the blade, as this is most important.
The motor pod is so designed mainly because it looks good. The streamlining probably
helps a bit, but is really unnecessary. On the other hand, it does provide a solid mass
behind the firewall (if you can call it that) and a little more weight. (The Cox 010
is awfully light.) The 1/8-in. plywood firewall is inlaid into the motor arm, and the
pod itself built up from the same 1/4-in. stock used for the blade, or anything else
in the scrap box, then carved and sanded to shape With the motor inverted, as shown in
the photos, the mounting screws bear through into the spruce arm instead of the soft
balsa. Add a wire guard loop if you think it necessary, but the prop generally seems
to be enough protection for the glow plug even when landing on bare spots.
 This has got to be the easiest helicopter
type thing yet-only one rotor blade!
 Rubber band mounted
weight boom affords easy adjustment of CG and crash protection.
 Inverted Cox .010
works much better than if it were upright, but no one knows why. Don't fly over concrete.
|
Thinned, clear, hot, fuel proof dope is used on everything except the motor pod and
adjacent portion of the motor arm, where straight dope is used for extra protection.
Colored dope could be used, of course, and should produce a pretty jazzy effect with
Charybdis revolving at the speed it does.
The inverted position of the Cox 010 has been found the best after much painful trial
and error. The tank outlet is slightly high, but the fuel line itself is at the extreme
outside of the swept circle, and fuel does not have to fight centrifugal force on its
way to the needle valve. The other arrangements that seem so obvious either don't work
for one reason or another, or offer no particular advantage.
Because of variations in engines, fuels, weather, altitude, etc., finding the proper
needle valve setting is something you must do yourself; there is nothing else for it.
But there is one peculiarity of the system worth mentioning that makes all the difference
in running time: an odd combination of forces and pressures is at work that requires
blocking off one of the filler nipples for maximum engine duration. With both nipples
open, engine run will be 15 to 20 seconds. With one blocked off, the run will be well
over a minute. This can be done with a short length of pinched-off tubing, or a longer
piece running to the pressure-tap nipple. With this latter system you don't lose the
short piece in the grass while fueling-it stays attached to the pressure-tap. You can,
of course, use a control-line tank and find the optimum setup yourself. However, the
shortest fuel line is always the best, and use of the integral tank has the advantages
of simplicity, strength, and aerodynamic cleanliness.
Flying
Launching may seem a bit hairy at first, but is really no problem if the CG is located
approximately as shown. To avoid losing fuel from the inverted tank, turn the Charybdis
upside-down and start as usual. Now grasp the hub area with your fingertips on the blade
leading edge and thumb on trailing. Raise the whole affair over your head while simultaneously
turning it upright; snap your wrist to start rotation, and push upwards. Then duck, and
run into the wind, since you probably haven't got the carburetion right to begin with.
Try again, until the engine continues to run and Charybdis climbs away like a startled
mallard.
An alternate launch method is to gusset the general area of the CG, drill a small
hole at the approximate center of rotation, and impale the Charybdis on a headless nail
driven into the end of a stick. In this case Charybdis will simply fly itself off once
it picks up sufficient speed. McCutchen remarks that this is a good idea while getting
the carburetion and balance unscrambled, since it prevents powered crashes while in an
unstable condition.
Once properly trimmed, Charybdis is remarkably stable. The rate of climb can be varied
by adding or removing clay, and by sliding the balance arm in and out. But this will
not make as much difference as you might think, and unless carried to extremes, will
not seriously disturb autorotation characteristics following engine shut-down. If the
Charybdis is really out of balance, of course it won't take off to begin with. Changing
the stabilizer angle, on the other hand, will make a great difference in rate of climb.
An adjustable stabilizer, or at least a movable tab on the' fixed one, would permit complete
freedom of experimentation.
The
other factor that most strongly affects performance is power. With the 3 x 1.25 standard
.010 prop, time from launch to touchdown averages around a minute and 30 to 45 seconds.
Charybdis climbs steadily for several hundred feet, depending on engine run, then descends
in autorotation for anything from 30 seconds to a minute. On one memorable flight of
this sort, the Charybdis got hung up in a thermal at 100 feet or so and just sat there,
silently autorotating over one spot for something close to two minutes. Total time was
3:15 for a rather different max flight.
But to really have a ball with Charybdis, try this: Turn the prop around to reduce
thrust, or use a 4-1/2 x 2, .020 prop, pile more clay on the balance arm, maybe try a
masking tape trim tab on the elevator-anything to hold it down. It will take a while
to work out, but you can get the Charybdis to hover waist-high. It will first sink to
a grass-cutting level, then find an equilibrium altitude in ground effect at two or three
feet, and just sit there, drifting with the breeze, following the contour of the ground.
McCutchen rigged one to do this so well that it would go down the gentle bank of a stream,
cross over, then climb the opposite bank and continue wandering off across the fields.
He says it was quite upsetting to casual observers along the flight path.
While hovering like this, Charybdis will produce the weirdest sound you have ever
heard outside a science-fiction movie. You'll think the Martians are coming-it is a kind
of whoop-whoop-whoop-whoop gradually increasing in pitch and frequency, with an underlying
humming note, and the scream of the engine. All this may be only a peculiarity of the
author's Charybdis, and may not be true for others-even working from the same plans,
everyone builds slightly differently. But for the Charybdis in the photos it's real.
And strange.
AAR Note: The Charybdis' principle of flight is very similar to that of the familiar
maple tree seed. Here is
NASA's page on them.
 <click for larger
version>
Notice:
The AMA Plans Service offers a full-size
version of many of the plans show here at a very reasonable cost. They will scale the plans any size for you. It is always
best to buy printed plans because my scanner versions often have distortions that can cause parts to fit poorly. Purchasing
plans also help to support the operation of the Academy of Model
Aeronautics - the #1 advocate for model aviation throughout the world. If the AMA no longer has this plan on file, I
will be glad to send you my higher resolution version.
Try my Scale Calculator for Model Airplane Plans.
Posted October 31, 2010
|