Well, firstly if you're interested in the theory behind it (why this occurs) I suggest you look up the physics of vortices in fluid mechanics. A fluid is usually a liquid OR a gas, and quite frequently in the cases we care about, they have similar qualitative properties. In this case, it's a form of chaotic or turbulent flow known as vortices. In this case, you are studying the effects of a vortex in air. The smoke just gives you a visual representation of what's happening (since the smoke tends to follow the air, and it isn't transparent to your eyes).

If you want to think of a vortex, think of a tornado or a whirlpool. That type of motion is what you have from vortices. In your case, imagine taking a long straight tornado (that has the same thickness at the top and bottom) and wrap it around into a circle. that's the type of motion the smoke ring is doing, and that's why it's so focused in a certain region of the air (like a tornado, in which you tend to only see it in a very focused region instead of spreading out in a larger region).

There are lots of names that this type of thing might be known as, such as ring vortex, or just plain vortices. I don't have a lot of good book references appropriate for the 7th grade level, but here's some internet links that might help you out.

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

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

I looked through those pages quickly, and they seem to give (qualitatively) the correct information (you can't always trust wikipedia!). But hopefully now that you know what you're dealing with, you might be able to find more resources on the internet, from your teachers, or at the library.

Hope this helps!

P. D. If you're more interested in the building process, I would suggest trial and error =). This is a relatively easy experiment to design, and you can try a couple of different things to see what kinds of effects you get. Say what if you use a longer soup can? Or a thicker one? What happens? These are all interesting questions. Is it better if the soup can is squarer? Etc. Does it have an effect or no real effect? What about volume of air inside the can? Different material (cellophane instead of balloon?) etc. It is perfectly legitimate to have the basic theory understood, and then try modifications to the basic design to attempt to find different qualitative results.

Here there are interesting resources for this question:

Websites:

click_1
click_2
click_3
click_4
click_5

Here's a video of two water vortex rings colliding!

video

The following link is to a chapter of a book that is at the college level (don't try to read it until you know vector calculus!)--but Figure 6.9 shows something interesting: a smoke particle in the smoke ring follows a loopy path.fluids

Real scientists are interested in vortex rings:

vortex_rings

and it's important to understand them for helecopters:

wikipedia
and
vortex_helic

Magazine article:"Ring Bubbles of Dolphins" by Ken Marten, Karim Shariff, Suchi Psarakos and Don J. White in "Scientific American", August 1996. (I haven't read this article, so I don't know how relevant it is.)

There is a book:Batchelor, G. K., (1967), An Introduction to Fluid Dynamics, Cambridge UP (reprinted 2000)(but this book might be at the college level; I haven't read it)

Good luck!

This sounds like a great idea for a project! So I think there are two things you must have control over with this. The first is exactly how much smoke you are putting in the can. This can easily be done by controlling how long you hold the burning incense inside the can, and just use a stop-watch to keep track. Second, you need to be able to consistently hit the balloon with the same force each time. To do this you can compress a spring to a certain measured length, and then let the side of the spring facing the balloon go, so that it hits the balloon. The farther you compress the spring, the harder it will hit the balloon. If the balloon is not too tight, you could pull the balloon back to a measured length, and then let it go to push the smoke out.

Other things you can vary are maybe 3) temperature or 4) air pressure measured with a barometer. Pressure would interesting but tougher to do,since you would have to perform the experiment at different elevations to get a different air pressure. Another thing 5) you can change would be the length of the can, by just gluing together more cans. Yet another variable 6) could be the diameter of the cylinder, by using coffee cans and buckets.This sounds like a great idea for a project! So I think there are two things you must have control over with this. The first is exactly how much smoke you are putting in the can. This can easily be done by controlling how long you hold the burning incense inside the can, and just use a stop-watch to keep track. Second, you need to be able to consistently hit the balloon with the same force each time. To do this you can compress a spring to a certain measured length, and then let the side of the spring facing the balloon go, so that it hits the balloon. The farther you compress the spring, the harder it will hit the balloon. If the balloon is not too tight, you could pull the balloon back to a measured length, and then let it go to push the smoke out.

Other things you can vary are maybe 3) temperature or 4) air pressure measured with a barometer. Pressure would interesting but tougher to do,since you would have to perform the experiment at different elevations to get a different air pressure. Another thing 5) you can change would be the length of the can, by just gluing together more cans. Yet another variable 6) could be the diameter of the cylinder, by using coffee cans and buckets.

Now, by varying one of the 6 things above and holding the others constant,you could see how this affects any number of things. You could measure how far the smoke rings travel before not becoming rings anymore. If the balloon hits the air fast enough, rings won't form anymore. Where is this threshold? There also should be a lower threshold too where slow enough speeds cannot make rings. What is this? It might be a little bit fuzzy about when you say it's a ring and when it isn't, but you should still get interesting numbers.

I hope this helps. Good Luck!