Rube Goldberg Machine--Junkyard Style : LUSENET : Junkyard Wars : One Thread

I propose a Rube Goldberg machine, done junkyard style. Each team would be given items for various steps of the challenge, and need to determine a creative way of connecting them to achieve a final goal. Certain items would also be required to be found from the junkyard and one used in each step. Once started, the team would not be allowed to touch the machine. The competition would be based on the following criteria: 1. Does it work? If one team's machine works start to finish, and the other's doesn't, they win. 2. How many of the predetermined steps were completed? If neither team's machine works completely, the team who acuated more of the items given to them wins. 3. How long did it run? If both team complete the challenge to the same point, or both complete it completely, the team who's machine runs longest without going over a prescibed time limit (10 min) wins. 4. If none of the above criteria can decide a winner, the team with the most complicated machine, as determined by the judge, wins.

Here is one setup that would work.

1. A quarter is inserted into a slot. 2. Using a wheel and axle from a motorized vehicle, close a switch to a small DC motor(given). 3. Using a sledge hammer from the junkyard, light a propane torch. 4. Using a compressed air cylinder from the junkyard, dispense a can of soda (given) down a prebuilt chute at the compitition site.

This provide enourmous potential for eduational sidetracks, the variety in potential construction is unlimited, and it does not require an expensive site for competition. The judge could be a mechanical engineer, preferrable with some association to R.G. competitions, and experts should definately have previous experience with R.G. competitions (many professors at engineering colleges have this expertise).

-- Erik L. Koski (, March 23, 2001


Yes, a really intersting idea - but VERY (almost totally) subjective in the judging. One judge could decide he likes the machine that is most reliable while the next decides he likes the one that is most outlandish. You wouldn't know what kind of thing to build and unless one machine just failed to work, then the whole thing would be a toss up. Really not a very good way to decide a winner...

-- Brian Flynn (, March 26, 2001.

I had a similar idea. Pick a common task for the machine like cracking open an egg into a pan. Require that each machine use the same startup, such as rolling a marble down a ramp. The machines could be judged with a combination of energy changes (the marble triggers a weight drop which sets off a ...) and time (longer is better). A simple formula could be used to determine a winner. For example: score = (# of energy changes) X (seconds of operation).

-- Paul Huff (, March 27, 2001.

u suck

-- blo joe (, April 15, 2004.

Moderation questions? read the FAQ