Saturday, December 09, 2006

amazing devices to impress your friends


Cool machines like bumblebees, that work but shouldn't. Prepare to test your intuition about nature. The more you know, the more surprising these will be.

1) Vortex Tube

This is a hollow tube of metal, somewhat like a flute, with one hole on the side and one on each end. It has no moving parts. You inject compressed air in the side, and the tube separates it into hot air which comes out one end and cold air which comes out the other. The hot can be boiling hot, and the cold can be so cold there is frost. There are no moving parts other than the air.


http://en.wikipedia.org/wiki/Vortex_tube
http://www.gyroscope.com/d.asp?product=VORTEXTUBE
also known as "Ranque-Hilsch vortex tubes"

2) The hydraulic ram pump

This is a hollow box with a hinged flap inside and three holes. You put it in a stream, and let water flow into one hole and out another. It causes water to pump itself out the third hole and up from the stream to your house on the hill. It has no other moving parts besides the water and a hinged flap, no motor, no outside power source.

http://en.wikipedia.org/wiki/Hydraulic_ram



3) the "siphon"

This is the familiar but mysterious hollow tube, open on both ends. You fill it with water, squeeze both ends shut, put one end in a high reservoir of water, put the other end lower than the other end, unclamp it, and water flows "by itself" up the siphon tube and back down and out the lower end. No motor or moving parts are required. Just the water flows.

http://en.wikipedia.org/wiki/Siphon


4) the Cavity Magnetron

This is one Western civilization has grown to love, and you probably own one even though you've never seen it.

The "cavity magnetron" is basically a block of metal with a fancy hole cut through it, a wire stuck down the middle,the air sucked out of the hole, a battery attached between wire and the sides, and the whole thing stuck between the poles of a magnet.

Turn it on and microwaves come out the end. It's used for radar and home microwave ovens. There are no amplifiers, electronics, vacuum tubes, or other moving parts.

http://en.wikipedia.org/wiki/Magnetron

5) Magic dice


Here's how to make a set of four dice with a little special twist.

Whoever picks second will win 2/3 of the time.

Here's the rules.

Each turn, you let your drinking buddy pick any of the four he likes.

You will pick one of the remaining three. You both roll the dice, and whoever gets the higher number wins.

Since you know the secret, you'll win most of the time.(You'll win two-thirds of the time, to be precise.)

After a while, he'll want the die you have, and you'll pick one of the two that you weren't using. You'll still win 2/3 of the time. He'll want yours, you pick one of the remaining 3, and you'll still win most of the time.

He'll probably cycle through all the dice at least 3 times before starting to punch you..

The secret is that there simply is no such thing as a "best" die. This is a fact that's true of most things, from cars to jobs to mates to politicians to sports teams, but it is so bizarre that we pretend it's not true. "How can there be no best one?" we always ask, always surprised.

Here's one way to make them:
Die A: (3,3,3,3,3,3) (a die with 3 on all 6 sides)
die B: (4,4,4,4,0,0) ( 4 sides with "4", two sides with zero)
die C: (5,5,5,1,1,1) ( 3 sides have "5", 3 sides have "1")
die D: (6,6,2,2,2,2) (2 sides with a 6, four with a 2)
It's probably best to make them different colors so they are easierto tell apart.

And here's the order they beat each other:

If he picks A, you pick B. Those fours beat the threes most of the time.
if he picks B, you pick C. If he picks C,
you pick. D.
And, amazingly, if he picks D, you pick A.

The dice form what is known as a "strange loop", and you can read about them in Martin Gardner's column in Scientific American where they are called "non-transitive dice." They really work.

But don't bet more than the price of the next beer, because you will almost certainly accused of cheating somehow. People generally can't accept the idea that there is no "best" one to pick. It drives them crazy.

6) "weak" and "strong" forces in nature

We don't realize that what you see depends on how big you are.

On the scale of humans, electric attractions are much "stronger" than gravitational ones. So, you can rub a balloon on your sleeve, charge it up, and it will stick to the wall, defying gravity.

But if you get out to the scale of the solar system or larger, electrical attraction more or less "goes away" and the force that determines the shape and fate of everything is gravity, the "weakest force".

Discussion from "Inquiring minds" about what "strong" really means.


============================

commentary:

These examples actually do relate to public health and aren't just party games.

When we're designing scientific tools or methods to detect "causality", the examples above can come in handy as part of our test cases.

These are important lessons for social engineering, whether you're designing an intervention, or trying to detect someone else trying to intervene in your life.

Sometimes the thing that determines the outcome is not at all obvious. In the examples of the vortex tube, or the magnetron, or the siphon, or even simple home plumbing, it is the large, fixed, static changeless structure that is determining what happens. All of the "obvious" interactions are just between adjacent electrons or molecules of air or water.

If we tried to "model" every water molecule in our city water supply, we'd overload our brains and computers. And yet, we can deal with the whole thing by raising up to bulk concepts, such as "volume" and "flow" and "pressure" that mean nothing at individual molecule levels, but mean something very real at the city water supply level.

Other lessons here: just because some effect or force totally dominates the world you see right around you doesn't tell you anything about which force or effect will win in the long run.

It is not obvious, without a great deal of thought, which small effects are actually "negligible" and can be thrown out, and which ones turn out to matter in the long run.

Essentially no scientist or superbright computer, that had never seen air before, would look at a beaker of air and "see" the possibility of tornadoes or hurricanes.

Those possibilities are"in there", but they're not, but they are. The dangers in multilevel modeling that discards feedback is that it also discards all the emergent effects that depend on feedback, such as the devices above, or tornadoes or hurricanes.

The bottom end of complexity is easy, the top end is easy, it's just the middle that's hard, but we don't need to go through the middle to get to the top end. It's just like plumbing - we just need concepts that are appropriate to the level we're thinking on.



(photo by Clearly Ambiguous )

technorati tags:, , , , , , , , , , , , , ,

No comments: