Flying Saucers

For the flight of the first saucer (28" dia), Blue Thunder motors were chosen
 to avoid typical white smoke and white flame seen in rocket pyramid flights.

The above saucer contains a motor mounting central spool consisting of a short airframe section, two bulkheads, and two canted motor mount tubes.  The outside skin is 1/2" building foam covered in fiberglass. Foam ribs arranged like spokes on a wheel help provide internal support and insure a consistent shape. 

Construction of the Small 28" Saucer Shown Above

The spool was constructed first including two slightly canted 29mm motor mount tubes for the 29/180 motors.  Next, two 29" disks of building foam and many ribs were cut.  The foam discs were lightly sanded to a taper on the inside edges near the perimeter to provide a flat joint.  The outside edge was lightly sanded to produce a slightly rounded edge.  Epoxy was applied to top and bottoms of the ribs and spool, and to the last inch of the insides of the building foam disks.  The foam disks were oriented so that their internal cell structure was the same direction for each.  (This was necessary because the foam is more rigid in one direction than the other and the stiffness had to match on the top and bottom.)  Center holes in the top, bottom, and spool were lined up and short pieces of masking tape were used to join the edges of the foam discs.   After the entire outside edge was taped, the epoxy was allowed to cure overnight.  The tape was carefully removed.  A layer of fiberglass was placed on the bottom of the saucer and overlapped around the perimeter about 6" onto the top.  This was taped to the top to hold it tight.  Much care was given to assuring a smooth equally distributed perimeter.  Epoxy was applied to the bottom and 1 1/2" over the top.  After curing, the tape was removed and the excess fiberglass was cut away being careful not to cut the foam.  The same process was then followed for the top of the saucer.  This provided an extra layer of fiberglass for the outside 1 1/2" on the perimeter of the dish.  A 1/8" thick disk of plywood was added to the bottom of the foam to protect it from the heat.  After the flight shown above, two nearly horizontal 24mm motor mount tubes were mounted into the bottom plywood disc to provide additional initial spin.

Background

For a horizontal saucer making a vertical flight, stability during powered flight was achievable as evidenced by spool rockets, pyramids, and Frisbee rockets.  These have been made as featherweight model rockets and large models have been built and flown with M motors.  None of these except for the monocopters utilized a spin.  For saucers, similar in design to Frisbee rockets, corrective force for stable flight is provided despite the unusual c.g./c.p. relationship.

Adequate spin on take-off was desired for my model to further insure a vertical ascent and to help stabilize descent. However, rotating solid saucer models tend to tilt slightly due to the wind and change attitude after burnout increasing to a steep angle on landing.  This allows the saucers to gain considerable velocity on descent.  (This resulted was slight damage to the edges of my first saucer design during several landings.)  Because of the drag associated with the saucer ascent, altitude is extremely limited with solid designs.  These solid saucers cease vertical ascent almost instantly when motor thrust ceases.

Fanblade based saucers are stable during ascent and descent and attain FAR greater altitudes.  But they can represent a hazard in larger sizes unless a guard screen is used.  These designs produce considerable lift which also produces more hang time near apogee and slows ascent.  Saucer rotation ceases and reverses 2/3 of the way back down to the ground.  Given sufficient rotation rate and vertical liftoff, these designs tend to land downwind.  Marlin Philyaw's fanblade rockets (like his Honda model below) appear to be the first amateur rockets of this type.  It is important for these models to achieve sufficient rotation prior to liftoff.  For this purpose, Marlin used two horizontally mounted Estes motors for the initial spin and two larger canted larger primary motors for lift. 

Extremely long shutter speeds capture an interesting spiral image as shown below in the photo of Marlin Philyaw's fanblade rocket.

 Larger fanblade saucer designs, created to maintain the illusion of the saucer shape, are involved and expensive to build because of the complexity of the turbine and required strength of the turbine for landings.  I began some work with fiberglass-covered solid balsa wingstock but have not completed it due to the disadvantage of constant prop pitch and the fanblade stress expected during landings.

My preliminary calculations indicated that rotating saucers under 5 ft in diameter could be build using fiberglass-covered 1/2" building foam sheets containing an internal spool for motor mounting and foam ribs.  Further calculations indicated that by using a structurally acceptable rotation rate and spacing the motors so that rotation will cover the distance between motors in 1/20 second, retinal persistence should allow the flame to be blurred enough to appear as a blue glow beneath the saucer.  Still cameras unfortunately do not capture that effect unless a low shutter speed is used and the saucer is followed by the camera.  This also blurs the saucer image as it ascends.  I believe however that done right, video can show the effect seen by observers of the flight.

Current Project 40" Saucer

Construction of a 40" diameter saucer is now complete.  It will fly on 4 - 38/480 motors (I300 blue propellant) providing a total of approximately 400 lbs of thrust.  The initial spin will be provided by 2 - 29/100 motors (also blue propellant).  The launch controller will ignite the spin motors.  When rpm is sufficient to provide over 1 g at the 7" spool internal diameter, the spin will close a centrifugal switch which will ignite the 4 larger lift motors.  Another centrifugal switch will energize 20 ultrabright LEDs on the underside of the saucer.  A timer will deploy parachutes slightly after apogee (5' chute for saucer, separate smaller chute for dome.)  The webbing will contain 2 separate swivels to allow the saucer to continue to spin during deployment and descent.  The saucer chute will be used to insure proper attitude during descent.

The central spool (shown in gray below) contains a 7 1/2" phenolic airframe with 2 layers of 8 oz kevlar/carbon fiber.  The 15" centering rings surrounding it are made of the new lightweight synthetic material shown on Rocketry online.  The bottom 1" of the spool contains an electronics compartment (with 3/16" plywood for the fixed bottom and removable top).  One inch sections of phenolic coupler are used inside the spool as spacing for the electronics compartment and to attach the dome.

The shell and dome are made of 1/2" building foam covered with 2 layers of 4 oz fiberglass (4 layers on edges).  Internal ribs of building foam are used to provide additional support.  (For clarity, spin motor mounts are not shown in the profile.)

The saucer weight is high.  For safety, the motor spool was overbuilt.  This made it heavy (abt 10 lbs including the 4 lbs of motors).  That has brought the total estimated weight of the saucer to 18 lbs.   Although most of the weight is concentrated in the center (15" dia), it still impacts the spin rate somewhat.  More importantly, landing weight without the dome and with the spent motors comes to 15.75 lb.  Without rotation, the motor mounts and large surface area of the base would handle the landing stresses.  With the heavier saucer spinning while landing (despite the combined drag of the saucer and the chute), the motor ends are somewhat more likely to damage the mount.  Hopefully, this will be mitigated by the surface of the spin motors which will be nearly horizontal.  The overall angular momentum on landing is also likely to be about 50-60%  higher than anticipated.

weight estimate:  18 lb   TOTAL

To create a complex changing light pattern, a LED controller was extracted from a $3.00 yo-yo available at Toys R Us.   Retinal persistence causes the switched LEDs to appear as dots and arcs as the saucer (or yo-yo) rotates.  Calculations based on the expected acceptable rpm for the saucer indicated that the number of rows of LEDs should be increased from 1 to 4 so that the saucer would only have to rotate 90 degrees (rather than 360 degrees like yo-yos and tops).  This also provides more light.  

The microscopic LEDs were cut away from the yo-yo LED controller and 5 transistors were added to drive a total of 20 ultrabright LEDs - 5 channels x 4 LEDs per channel.  I used IC sockets like the one shown to the right of the board so that I can change the LED current limit resistors if the number of LEDs installed is altered or the LEDs are replaced with LEDs of different colors.  (LEDs are mounted recessed in Christmas tree light sockets on the saucer to facilitate their replacement in case a change in the color pattern is desired.) 

After many tests, the LEDs grew dim.  This was a result of draining the small hearing aid batteries which drive the transistors.  So, their holders were removed and they too were replaced with a set of AAA batteries.  This restored the LEDs to full brilliance.

UltraBright LEDs useful for Saucer

I'm estimating an altitude of about 300 ft.  The divergence angle for the light from the LEDs is much too narrow for daylight viewing at required launch distances.  Experiments with clear plastic jewels for covers, sanding the LEDs to diffuse them, and using translucent covers all produced little improvement.  Producing divergence by sanding the LEDs flat and polishing them caused an unacceptible drop in intensity for daytime operation. 

However, I spun the saucer up for a test at night and the effect is spectacular and visible at an angle even without being directly in the primary beam path....  ;-)

Currently the 5 LED channels are arranged from the outside to the inside as blue, red, yellow, yellow.   The blue is the color you see under cars.  Cameras with short shutter speeds freeze the motor plumes in place rather than blending them like your eye does during the rotation.  Still cameras therefore won't capture the appearance of the flight accurately unless a composite of high and low speed shots is made to capture the saucer and the flame separately or the flight is captured with a low shutter speed immediately at liftoff before it achieves significant vertical velocity.

So, what was the 40" saucer flight expected to look like?

The appearance was expected to bear at least a faint resemblance to one of the following drawings:

The saucer has been painted with high intensity long life (8 hours glow time) strontium aluminate- based  "aqua" glow-powder and glows an eerie blue-green color.  That should make for a very interesting night launch.

Tiny Amarillo Blue spin motors clearly visble here quickly
brought the dish to a little over 420 rpm in the spin test video.

Current Project FLIGHT

The saucer looks much bluer at night and very light yellow/green during the daytime.

The flight was amazing.  With the exception of the chute inflation, ALL systems worked as planned.  The saucer sound was in sync. with the flight.  The spin motors lit and provided the desired spin rate for the saucer.  The centrifugal switch lit all four lift motors on cue.  The flight was perfectly stable and vertical with a bright blue column of energy which (because of the spin rate) appeared to merge into a large diameter blue area with a point.  Surprisingly the glowing saucer shape was clearly visible even with the motors lit. And the saucer lights created geometric patterns also visible during and after the motor burn.  The altitude was just as expected.  And, the timer also put the parachutes out right on cue.

As one observer put it "....flying saucer in the traditional 1950s shape complete with flashing lights and a 1950s movie sound track. In addition to the flashing lights, the entire saucer glowed because I believe it had been painted with glow in the dark paint and charged up in the sun all day."

Another commented, "As the saucer lifted, the blue thunders gave off the effect of a solid but thick blue ring under the saucer.  It looked like some kind of advanced propulsion system just like you see in the movies!  It was really cool!"

IMAGES FROM VIDEO OF FLIGHT

The appearance of the actual flight was FAR DIFFERENT than the video.
As expected, it looked like a combination of all the concept images and spin test.
  However, the unexpectedly bright and intense blue of the motor plume was captured
 on the video as bright white. The clearly defined glowing saucer and lights were visible 
to the eye before and during the launch but are completely invisible on video.

capture 1 (instant lift motors ignition)	capture 2	capture 3
capture 4 (estimated peak flame diameter 30")	capture 5	capture 6 (looking up somewhat later)

And, below are two captures from Tony Hewit's video of the flight.

Below is a sample created by turning the brightness down on the Tony's image without changing the color.
It revealed the following sample around the flame.  During flight, the flame was saturated and very bright.

 I don't know how to reproduce a solid extremely bright intense blue in the brightest white part of a drawing.
But what we actually saw during the launch was closer to the drawings below than to the videos.

We heard the ejection event on cue but as the saucer hovered and began to descend slowly, the cargo chute used on the main saucer body did not inflate.  The saucer did not accelerate rapidly toward the ground until it tipped over on its side about 2/3 of the way to the ground.  Then from such a low altitude, it hit before chute inflation and was damaged.  Inspection of the webbing and chute revealed no tangling.  The webbing was laid out with NO twists in it.  The 2 barrel swivels at different locations worked perfectly.  Surprisingly, when we recovered the glowing saucer, the LEDs were still lit and flashing with the 2 AAA, 3 AAA and 2 9V batteries, LED processor, and ejection timer all in place and functioning.  It appears that the only weakness was the choice of chutes which require more airspeed.  Next time, cheap thin chutes with light shroud lines will be used.  Normally, a bad recovery is a bad flight.  In this instance, a severely limited splash pattern with sufficient observer distance provided the planned safety margin.  And despite the desire for recovery,  the flight itself and proofs of all the varied special effects concepts and design approaches was the primary goal.  With so many systems performing beautifully and the the special effects creating the force field illusion so beautifully, the project was a resounding success despite the damage.

Flight Damage

When we picked up the saucer, there were glowing blue/green specs all over the ground around it.  Here's the saucer the next day with the motors removed.  

Conclusions Regarding Solid Spinning Saucers

• Ground-based spin motor ignition followed by centrifugal-switch-based lift motor ignition using low current igniters and ni-cad batteries works.
• Because of its almost instantaneous and reliable ignition characteristics, fast blue propellant such as Blue Thunder is an excellent choice for saucers.
• A launch rod with a length of half the diameter of the saucer is sufficient and much longer rods are undesirable.
• At high spin rates with widely spaced motors, the saucers are still very stable and produce huge exhaust plumes which retain their brightness.
• These saucers tend launch vertically and descend downwind (and somewhat to the left or right based on direction of spin) unless a chute is deployed.
• At 420 rpm, rapidly blinking lights can be set up to merge into a geometric pattern clearly visible during the flight.
• Glow-in-the-dark saucers can be seen during night launches but tend to be invisible to cameras.
• Saucers reach apogee at the time simple diameter-based simulations predict.  Therefore, timers are excellent deployment solutions.
• Use of 2 swivels (one placed high and placed low) on the webbing is very effective in preventing winding of the webbing and chute.
• Normal launch distances are appropriate because of the low altitudes and restricted splash pattern.
• I believe that recovery requires low porosity chutes which open immediately even at extremely low velocities.

NOTE:  Daytime flights would not be likely to produce the force field illusion
             created in this project and would make the LEDs very difficult to see.

With so many requests for another flight, the present plan is to prove the recovery system with small test saucer flights and to build the next generation saucer.  (Of course the other "secret" Hollywood-style special effects project I have in mind is an 8 ft diameter x 12 ft tall starship similar in style to the one in "The Last Starfighter" or the fighters in "Battlestar Galactica".  I'm not certain which project will win out ;-)

Simpler featherweight saucer rockets are commonly available:

Estes Snitch                                             OOP Quest saucer

For those wishing to build featherweight recovery saucers, Art Applewhite produces kits which have been flown on a variety of motors, are inexpensive, and are easy to build.  Below are a few examples.

The Wocket by Skunk Works is a fiberglass shell kit with a wooden central structure and features rear ejection.  Their 24" model flies on 54mm motors.

And, for grins, check out the following edible saucer.  It certainly brought a smile to the kid in me.
http://www.vatsaas.org/rtv/misc/oddrocs/toastite.aspx