Jet Propulsion Laboratory National Aeronautics and Space Administration California Institute of Technology
NASA Logo - Jet Propulsion Laboratory + Visit JPL Education Gateway
+ Visit NASA Education Portal
JPL Home Earth Solar System STARS AND GALAXIES Technology
Education Gateway
Education Gateway Home Page Educators Students Educator Resource Center Informal Education Team Competitions JPL Education Kids
   + IC HOME
   + IC ARCHIVES
2007 Results
2006 Results
2005 Results
2004 Results
2003 Results
2002 Results
2001 Results
2000 Results
1999 Results
1998 Results
JPL Annual Invention Challenge - 1999 Gallery
 
 
Welcome to the 1999 Annual Invention Challenge's Invention Gallery. Clicking on the titles will provide notes from each contestant about what they invented, how they made it, and lessons learned from their invention. There is also a VHS-format video tape of the 1999 Invention Challenge which can be borrowed by emailing your request to Paul MacNeal at Paul.D.MacNeal@jpl.nasa.gov.


Participants | Contest Overview | Background | Invention Gallery | Contest Rules


Winners, etc.
Trophies
Browse Image (35 KB)
Trophies
Browse Image (42 KB)
Trophies
Browse Image (40 KB)
Winners
Browse Image (58 KB)
Winners
Browse Image (59 KB)
Winners
Browse Image (44 KB)
Winners
Browse Image (53 KB)
Winners
Browse Image (44 KB)
Winners
Browse Image (49 KB)
General
Browse Image (32 KB)
General
Browse Image (40 KB)
General
Browse Image (30 KB)
General
Browse Image (66 KB)
General
Browse Image (68 KB)
General
Browse Image (36 KB)

"Right On Baby"
Closest & Most Artistic
Right On Baby
Browse Image (50 KB)
Right On Baby
Browse Image (56 KB)
Right On Baby
Browse Image (70 KB)
Right On Baby
Browse Image (41 KB)
Right On Baby
Browse Image (40 KB)
Right On Baby
Browse Image (44 KB)
Right On Baby
Browse Image (46 KB)
Right On Baby
Browse Image (46 KB)
 
My entree - "Right On Baby" (Pronounced with an English accent like Austin Powers) was a simple catapult design. The basic design criterion was make the device stiff and repeatable. The design material was 1/2" copper plumbing soldered together.

The release angle of 45 degrees was chosen for minimum velocity and thus the minimum drag coefficient. The longest and strongest screen door spring was chosen as a starting point and was retained throughout. The connecting cord is clothesline, which is extremely flexible but doesn’t stretch. Hasps connect each end of the cord providing a non-slip connection. Two ball bearing pulleys guide the cord between the reaction arm and the spring.

Finally an adjustable tensioner is fixed to the spring allowing the (spring tension) distance to be adjusted. The ball holder is a strainer which allows the air to get behind the ball during release. A double sided copper board bored with a hole large enough to let the ball through and 1/8" copper tubing reinforcement was added to stiffen the strainer. A sighter removed from a telescope is used to correct for miss-alignments. The ball is fired and the sighter is adjusted to point to where the ball landed. Once adjusted the sighter will point in the general direction of the balls path.

After figuring out that the strainer was flexing throwing the ball down instead of up, re-enforcements were added. The reaction arm stop is a 1/2" rubber stop. Every time the reaction arm slams into the stop it jars loose the setscrews on the scope. This limits the accuracy to about two shots before re-calibration of the scope. I believe changing the setscrews on the scope would increase the repeatability.

"The Jolly Parcel Launcher"
2nd Closest
The Jolly Parcel Launcher
Browse Image (45 KB)
The Jolly Parcel Launcher
Browse Image (44 KB)
The Jolly Parcel Launcher
Browse Image (42 KB)
The "Jolly Parcel Launcher" is an ordinary catapult powered by a rat trap. It is made of two-by-four segments screwed to a plywood base and counter-balanced by a pair of cconcrete blocks. Tin from the bottom of a coffe can, and wired to the rat trap, formed the payload holder. A laser pointer was used for azimuthal aiming. Distance was programmed into the JPL by pulling back the rat trap a measured amount.

The largest source of variability was the friction, caused by centrifugal force, at the instants of launch. Differing distributions of the beans in the beanbag produce differing amounts of friction. This effect was reduced by bending the holder back about 20 degrees.

If given another shot, I would weigh the beanbag and pad the JPL with pennys until a standard weight was reached.

"Li'l Stinker"
3rd Closest
Li'l Stinker
Browse Image (41 KB)
 
The device was a Spring Loaded System. It consisted of a large metal spring, PVC piping, a rod to hold the plastic cup that held the hacky sack, and the wood base. I'll send you the "drawings" when I get into work on Monday. Challenges - calibrating the spring to get the proper distance. Improvement - refining the targeting capability.

"K'nexTM Katapult"
Lightest & Smallest
K'nex Katapult
Browse Image (66 KB)
K'nex Katapult
Browse Image (67 KB)
K'nex Katapult
Browse Image (48 KB)
K'nex Katapult by Alan DeVault
Browse Image (45 KB)
 
My entry, the K'nexTM Katapult, was my attempt at a "Faster, Cheaper, Better" entry.

It certainly had to be fast because, although I had the idea of what I planned to do from shortly after the challenge was announced, I didn't actually start working on it until two nights before the contest.

It was definitely cheap; since it was built entirely with parts already on hand, I didn't have to spend any money on it.

Whether it was better or not is debatable. But, in spite of a bit of bad luck on the official toss, resulting in a disappointing performance, it worked well in practice launchings, and proved to be quite accurate and repeatable. Also it did manage to get the award for the smallest and lightest entry.

The structure was made entirely from parts from my son's K'nexTM toy building set. This set consists of plastic struts and connectors which snap together to build surprisingly strong and rigid structures. The ease with which this allowed the structure to be quickly built and changed, permitted a quick cycle of testing and modification to refine the design.

Propulsion power was provided by a hand-full of rubber bands grabbed from the butter dish in the kitchen drawer, where they get thrown after being pulled off of the morning newspaper. The throwing range was adjusted by adding and subtracting rubber bands until the desired range was achieved.

In test launchings, the landing position proved to be consistent within 4 to 6 inches. Just prior to the contest, a quick test in the hallway outside my office showed the range to be a few inches too long. Rather than go through a last minute round of adding and subtracting rubber bands to fine tune the range, I elected to just position it farther back in the launching zone to compensate.

My practice footbag was very close in weight to the one of the two official contest bags, which I chose to use for my official toss. So I expected to be pretty close to the target. I was very surprised when my official toss was so far off.

On examination of the 'katapult' after the contest, I discovered that, unfortunately, one of the struts that acted as the stop for the launching arm had come loose from its connector at one end. This allowed the arm to be over-cocked and slightly twisted, resulting in the inaccurate toss.

After the contest was over, I corrected this, and made a couple of unofficial tosses from the official launch platform, and the resulting shots were right in there with the top finishers. Unfortunately, only the one official toss counted.

"Pandora's Box"
Most Unusual & Largest
Pandora's Box
Browse Image (46 KB)
Pandora's Box
Browse Image (58 KB)
Pandora's Box
Browse Image (54 KB)
Pandora's Box
Browse Image (55 KB)
Pandora's Box
Browse Image (46 KB)
Pandora's Box
Browse Image (47 KB)
Pandora's Box
Browse Image (62 KB)
Pandora's Box
Browse Image (23 KB)
Pandora's Box
Browse Image (58 KB)
Pandora's Box
Browse Image (58 KB)
Pandora's Box
Browse Image (71 KB)
 
Pandora's box combined two concepts: 1. A large helium balloon with a slow leak to lift the foot bag and then set it down and 2. A lenticular strip to position the foot bag over the target. (The use of a lenticular strip, similar to that in a metal tape measure, was suggested by Don Moore.) The lenticular strip was made by taping aluminum (light weight) Venetian blind slats together, so as to form a flat hollow tube, which maximized the radius of gyration in the direction normal to the slat surface.

Unfortunately, the aluminum used in Venetian blinds is very soft, so buckling of the lenticular was still a problem. The buckling problem was solved by using six oblong "jelly bean" balloons taped end-to-end on top of the lenticular. (The six oblong balloons were approximately 24" in length and 10" in diameter and also filled with helium.) The most difficult challenge was fitting the six oblong balloons into the allowed volume, a box 30 cm on a side, and routing the lenticular between the adjacent balloons.

In order to facilitate this, it was necessary to have a small helium tank inside the box to inflate the large (40" diameter) balloon during the launch sequence. Even so, it was impossible to close the lid of the box completely on the morning of the contest, because the six oblong balloons were erroneously over inflated. (This inability to close the lid prevented the use of an automatic system designed to release the large balloon from the helium tank and to open the box for the six lenticular balloons to escape.)

A lesson learned follows from when during the balloons descent, I illegally, and unwisely, fanned the big balloon to move it to the south; soon after which the foot bag touched down a meter or so too far to the south.

"Leave well enough alone."

"Pipeworks"
Most Creative
Pipeworks
Browse Image (32 KB)
Pipeworks
Browse Image (49 KB)
Pipeworks
Browse Image (55 KB)
Pipeworks
Browse Image (61 KB)
Pipeworks
Browse Image (48 KB)
Pipeworks
Browse Image (58 KB)
Pipeworks
Browse Image (61 KB)
Pipeworks
Browse Image (43 KB)
Pipeworks
Browse Image (42 KB)
How It Was Supposed to Work

The Pipeworks entry was composed of 15 concentric telescoping tubes. Each tube was about 17 inches long with two inches of overlap to provide rigidity. Each tube had the end closed off to provide rigidity and a lateral support for the tape measure. A 25 foot long tape measure ran up the inside of the tubes in order to provide the compressive thrust necessary to extend all the tubes. The tape measure was sandwiched between two spring-loaded rollers. One of the rollers was attached to a 120 Volt motor capable of 15 inch-pounds output torque at an operating speed of 32 RPM. The entire assembly was mounted to a plywood base that held the assembly at a 45 degree angle to horizontal. In theory, once the electrical switch was turned on, the motor was supposed to drive the tape measure out the full five meters (in approximately 30 seconds), at which time the fishing line would become taught and release the footbag exactly on the target.

The Competition

Early in the construction of the device, I found that the "slop" of each of the joints was causing a tremendous curvature to my originally planned straight line. By making each of the tubes with less "slop", the force required to move the tubes became too great, so I had to sacrifice "slop" for ease of deployment. To make a long story short, I knew before the contest that the best I could do was about 12 feet (3.65 meters), because the force required to push a curved set of tubes was greater than the capability of the tape measure in compression (i.e. it buckled prematurely). On the day of the contest, I had a structural failure of a bonded joint that was intended to deliver the power from the motor to the rollers. Unfortunately, it occurred after only deploying about 9 inches. After I manually pulled the tubing out about 14 feet, the entire tube structure broke near the base and fell to the ground. Needless to say, everyone had a good laugh (including myself).

To Make It Better

To improve the design, it would be necessary to use better (and much more expensive) materials, such as aluminum, for the tubing. This would enable the tube to remain much more straight and yet still provide a relatively light, friction-free assembly. The weak link to the design was the measuring tape. Perhaps using two tapes in parallel could have provided more capability.

"The Big Sucker"
The Big Sucker
Browse Image (48 KB)
Big Sucker
Browse Image (46 KB)
 
This particular invention took hours to develop. (I think I actually invested all of 2 or 3 hours, including the contest itself.) I had wanted to create some fancy spinning motor, computer-controlled, wind sensing catapult, but work and personal schedules took most of my time. Though I was the first person to enter the contest, I thought I'd have to back out … until the day before the event. Then while slumbering on my bed, I got the idea to make an air powered cannon, by using my shop vacuum (i.e. The Sucker).

My vacuum doubles as a leaf blower, so my first idea was to stick the footbag in the bottom of the blower tube, and turn on the motor and measure the distance. At first, nothing seemed to happen. Then after many seconds, the footbag slowly began to peer out the end of the tube. A mild tap on the side caused it to pop out, and land about 8 feet away. I reasoned I must need a larger more powerful vacuum (i.e. The Big Sucker). While bringing my leaf blower tube into work (to try on my boss's shop vacuum) I noticed that one end of the tube was NOT stepped-out to accommodate the attachment of a hose, but instead the entire tube was conical in shape, with the Outside Diameter of one end matching the Inside Diameter of the other. Suddenly I realized the footbag was being compressed as the air pushed it further up the tube. No wonder it barely popped out!

Thankfully the Big Sucker was configured to accept the narrower end of my tube on its exhaust port, allowing the footbag to feel less and less pressure as it traveled up up and away. That day at lunch I set up the vacuum in the parking lot with a couple blocks under the wheels to angle the tube forward. As luck would have it, my footbag landed the perfect distance away, and exhibited fair consistency. Of course, that was with MY footbag.

At the event, the official footbag appeared to be slightly larger, and was quite stiff as I jammed it into the ejection tube. This caused more pressure to be built up, which resulted in a stronger launch than during my extensive prior testing. Alas, I overshot the target, but had a lot of fun doing it!

"The Old Tosser"
The Old Tosser
Browse Image (56 KB)
The Old Tosser
Browse Image (68 KB)
The Old Tosser
Browse Image (40 KB)
The Old Tosser is a spring loaded catapult; by pulling back on the rusty spoon the bmx bicycle forks hinged on the tricycle handlebars extend the springs which are attatched to the pedals of the trike. The materials used are all discarded metal items collected from the alleys and dumpsters of the boss basin. The creation of a new and useful thing out of the ditritus of society is a 'hobby' of mine.

As is said: one man's trash is another man's treasure. Ahh, the challenge, have you ever seen a piece of rusty junk by the side of the road that wasn't all bent out of shape ?, neither have I. Which leads right to the problems encountered, is anything square ?, isn't square just something we made up anyway ?. I'll admit square is easier, but where is the challenge in that ? the major improvement I would have made would be to the performance itself; we really wanted to win.

I suppose the sighting devices that assisted the first and second place finishers were a good idea, but hey, loose variables can be fun too. Thanks for the opportunity to play, we had a great time and look forward to next years challenge.

"The Portuguese Chowder Society"
The Portuguese Chowder Society
Browse Image (32 KB)
The Portuguese Chowder Society
Browse Image (50 KB)
The Portuguese Chowder Society
Browse Image (57 KB)

"The Mouse-apult"
The Mouse-apult
Browse Image (51 KB)
The Mouse-apult
Browse Image (69 KB)
 
Creators Christian Carlberg and Eric Gill made Mouse-a-pult with three things in mind:

- It had to be weight powered
- It had to be adjustable
- It had to throw the bean bag in a high arc

Gravity
We were afraid that springs or bungee might change significantly enough to lose consistency, and consistency was the key to winning. We powered our thrower with weight and relied on the constant of acceleration.

Adjustability
We built a lot of adjustments into the M.A.P. The throwing arm could be lengthened or shortened, and the start and release points were also adjustable so we could tune in to our target.

Archwork
The rules stated that, incase of a tie, the winner would be the one that rolls the least. With this in mind we set M.A.P to throw at a high arc and thus have little roll after impacting the ground.

All in all, we had a great time building and demonstrating our thrower. M.A.P. is now retired but from time to time is called into service to deliver food or generally annoy co-workers.

"The Five Meter Flingster"
The Five Meter Flingster
Browse Image (41 KB)
The Five Meter Flingster
Browse Image (45 KB)
The Five Meter Flingster
Browse Image (38 KB)
Ours was a last-minute entry, with the design based directly on a toy catapult which one of our team members owned. As we tested the toy, we discovered that rubber bands would be capable of launching a footbag the desired distance. So we merely built a large-scale model of the toy. The wooden launching arm was attached with a standard cabinet hinge. It was surprisingly consistent, despite the play evident in the hinge. Before firing, the arm was latched between two wood blocks, to assist in side-by-side alignment.

As testing progressed, we learned that rubber bands relax over time. We replaced the bands with multiple springs. The springs could be adjusted up or down the arm to increase or decrease their launching force. The stop for the launching arm was adjustable (pin-and-hole) for coarse adjustments on the distance traveled. The footbag sat in a wooden cup which could slide up or down the launching arm in a slot, for fine adjustments on distance. Once the correct distance was achieved, the cup was fixed in place with a screw and nut. Aiming was done using two pairs of aligned sighting sticks; one on each side of the launching arm. The best part of the Flingster was its precision. It hit the target distance very well. The biggest problem was its aim. It must have shifted during the 30-minute drive to JPL, making the preset sights incorrect. We should have tested it after our arrival at the Lab, but before the contest began.

"Coin Toss"
Coin Toss
Browse Image (43 KB)
Coin Toss
Browse Image (43 KB)
 
Coin Toss is a simple weight driven catapult arm. My goal was to make it repeatable. My original idea was to use the diagonal length of the 75 cm space to get more length. As it turned out I didn't need to resort to this, for when the device was in launch position it was less than 75 cm long. The main body of the device was a wooden base with a wooden tower.

The catapult arm was 2 inch PVC pipe that was augmented with nylon bushings to provide smooth rotation. I used a plastic cup to carry the beanbag and trimmed it down until it released the bag smoothly (and consistently). Both the plastic cup and the rope that held the weight were attached to the pipe with hose clamps that provided a secure attachment, but were easy to adjust as the plan came together.

My biggest problem was getting a significant mass that was small enough and robust enough for continual dropping. I decided that coins were a heavy easily obtained item that could be packed in dense manner. I found that an empty milk jug was a fairly compact but sturdy container for the coins. As it turned out, I ran out of coins and needed to find more weight. I found a container of nuts and bolts and then added some sand as I fine-tuned the distance that the beanbag was thrown.

"Troy Warrior Flinger"
Troy Warrior Flinger
Browse Image (37 KB)
Troy Warrior Flinger
Browse Image (40 KB)
Troy Warrior Flinger
Browse Image (83 KB)
This father-son entry is named the Troy Warrior Flinger after the son's high school (Troy High School, Fullerton, CA; mascot = Warriors). We used a basic sling-shot design. Our apparatus had two parallel arms about 24" long and separated by about 10", which were inclined upwards at an angle of about 50 degrees. The energy was supplied by two heavy-duty earthquake springs which were attached to the ends of the arms. Since precision was the primary objective of this competition, and there was no limit on weight, we used very heavy-duty construction. The main structural components of the apparatus were made of heavy metal Unistrut, and they were assembled by large bolts. During assembly, we joked that we could lift the engine out of a car with these materials. Our finished framework had extremely high rigidity, and in test trials, we were able to achieve remarkable reproducibility.

We mounted a holder on the lower end of the springs to hold the footbag. Some of our early designs used a holder which was significantly larger than the footbag, however we found this to be inaccurate. We eventually settled for a holder made out of the bottom of a plastic water bottle, which was the same size as the footbag. Our calibration system consisted of a ruler taped to the left arm, which measured how far the springs had been stretched. After several days of calibration we discovered that we needed to extend the springs exactly 3 ˝ inches (and the footbag flight distance was sensitive to variations in this measurement of 1/16 of an inch). In our early designs, we had problems pulling the holder down exactly the same distance every time. We corrected this problem by attaching to the holder a fingernail file with one end cut into a point. This allowed us to read the ruler much more precisely.

Our trigger consisted of a bolt through which we had drilled a hole in such a manner that we were able to thread a string through it. Then when we turned the bolt it pulled the holder down. To release the device we cut the string with a pair of wire-cutters. We started out with an upwards angle of about 60 degrees to keep the Flinger from tipping over and to minimize the roll. In our final design we used an angle of about 50 degrees. To aim the Flinger, we added a gun-sight on the right side consisting of two vertical bolts that were aligned by trial and error. We decorated one of these bolts with a JPL antenna ball.

In test trials we were able to repeatedly drop the footbag into a cup, which indicates precision of about 5 cm. The device tended to be more precise with its distance than with its azimuth. This ended up being our flaw because in the competition our throw was almost exactly 5 meters but it was out of alignment.

"ACME Footbag Catapult"
ACME Footbag Catapult
Browse Image (37 KB)
ACME Footbag Catapult
Browse Image (36 KB)
ACME Footbag Catapult
Browse Image (55 KB)
ACME Footbag Catapult
Browse Image (58 KB)

Browse Image (64 KB)
 
Building materials:
Springs (2), Plywood, 2X4 pinewood, screws, nails, paint, bolt/nut/washers
Operation of the device:
Your basic catapult, lever pulled down by hand to a pre-determined position (based on several trial runs), manual release, cross fingers for luck!
Most difficult challenges:
John Schofield and I submitted companion entries. We developed both designs independently of one another during construction, but established the basics by shotgunning the approach.
Our original idea was a telescoping series of tubes (similar to your design, Paul) which would position the footbag right on the target, but we could not find materials that were strong enough while being light weight. We wound up settling on a catapult type of structure.
The main challenges were:
Use one spring, two springs of what length and spring rate, length of the lever, how to launch the bag in the air without it slipping down the lever, etc.
Improvements that could have been made:
Use of lighter weight materials, reduce spring bind, bolt the machine to the plate provided, incorporate aiming device.
 
 
 
NASA Privacy Statement Glossary Sitemap Feedback
FIRST GOV   National Aeronautics and Space Administration