Starting several years ago, when my son was in kindergarten, I’ve made a number of very simple launchers for soda-bottle rockets. It started with a little talk and demonstration of pressure I gave to my son’s kindergarten class. I had several small activities, but the most impressive was a soda-bottle water rocket. Here are some plans I wrote up for the launcher: (PDF format) (instructions in Spanish). The design of the launcher is simpler than many, as it uses just a friction fit to hold the bottle in place, and the rocket launches as soon as the pressure gets high enough, rather than using a more complicated retention and trigger mechanism. This does add an element of unpredictability to when the rocket launches and makes it harder to do experiments that involve adjusting the pressure precisely, but the launcher is very easy to make and the relatively low pressure on release prevents bottles from exploding.
I started with an in-class circle talking about pressure. I had each student feel the air pressure against their fingers from a bike pump, then showed them water shooting out small holes in the side of a bottle (noticing how far the water went for different amounts of water above the hole), then had them get in a circle holding hands (the bottle), and had some of them play the role of water molecules bumping against each other and the “bottle”. After that we went outside to launch the rockets. During the next week the kids decorated their own rockets with stickers and markers, and we had another launching where each child launched his or her own rocket. It was great fun, and the activities could easily be adapted to older students (by having them do some actual experiments).
For Fiesta de los Artes at the same elementary school when my son was in second grade, I set up a rocket-decorating table with 5 rocket launchers—it was a very popular activity. Using colored (and black) vinyl electrical tape proved to be the most popular decorations, though shiny metallic paper with adhesive backing was also popular. Lick-and-stick stickers were not very popular, but 1-cent kestrel stamps were moderately popular, and much cheaper than commercial stickers.
I did this project again for my son’s 5th grade class at a different school (not including decorating the rockets), where it was again very popular. We spent some time before the launching discussing Newton’s laws and how rockets worked, and some time afterwards discussing what one could measure on the rocket launches and how to test various ideas about increasing the height rockets went to. In after-school tech club, we had a session where students could build their own rocket launchers (for the cost of the PVC parts)—this was also quite popular, even with the younger kids. For more information about experiments to do with rockets see http://exploration.grc.nasa.gov/education/rocket/TRCRocket/RocketActivitiesHome2.html , particularly http://exploration.grc.nasa.gov/education/rocket/BottleRocket/index.htm
I still have two of these launchers in my garage, but I haven’t launched a soda bottle for a couple of years. After seeing the post Threaded Lamp Rod Process, I’m strongly tempted to try again. The idea there is to join two 2-liter bottles bottom-to-bottom with threaded lamp rod to make a 4-liter rocket. Will the bigger rockets be too heavy for this launcher? (The T-joint unscrews to make the launcher easier to carry on a bicycle.) Will the bigger rocket need more pressure, and hence a positive retention system and trigger release?
I wonder if I can work soda-bottle rockets into the calculus-based physics class we’re doing as home-school this year. They are certainly a nice demonstration of conservation of momentum. I don’t know whether we can make adequate computational models of the rocket though, as I think that the exit speed of the water depends as much on frictional forces that are hard to model as it does on pressure and nozzle area. If we can get good enough video of a launch we might be able to estimate how long it takes to empty the water from the bottle and how fast the bottle is moving, which may be enough to do some decent modeling.