This site describes homemade water rockets constructed from plastic pop bottles. The bottles are one of those rare miracles of serendipity: although designed for something else altogether, they are nearly perfect for making water rockets. They're designed to hold liquids at high pressures, they're very lightweight, and they have a conveniently located ring molded around the neck that's handy for holding the bottle down while pressurizing it. About the only thing that could make them better is to have the fins and nose cone molded in for you.
The rockets made from these bottles are surprisingly powerful. A standard 2-liter pop bottle 1/3 full of water, pumped to 80 psi and then released, will eject all its water in less than one-tenth of a second, and at that point ("burnout") will be only about 2 meters off the ground. Amazingly, its velocity at burnout is around 76 meters per second. That's over 170 miles per hour! This means the average acceleration during thrust is 111 g's! Yowza. Safety rule number 1: Never get in the way of one of these rockets...
A plain bottle is of course aerodynamically hopeless, and tumbles fluffily after burnout, rarely going higher than 60 or 70 feet. But if you add fins and nose weight, transforming the bottle into an aerodynamically stable rocket (see rocket theory), then all that initial thrust can be transformed into truly amazing altitudes of over 300 feet. You really have to see it to believe it.
For me it all started with a small plastic launcher that used to be sold by that wonderful place that every hobbyist should know about, American Science and Surplus (once known as Jerryco), for around 10 dollars. The same launcher, called a "POP FLIGHT" launcher, is also sold by Apogee Components, a model rocket supplier. The catalog copy said I could make a water rocket from a plastic pop bottle: that sounded like a fun little diversion, so I ordered the thing on a whim. When it arrived, I innocently took it out to my tiny San Francisco backyard, put some water in a 2 liter bottle, clamped it on the launcher, pumped it up to 40 psi or so, and yanked the string. I wasn't quite ready for what happened next...
The explosive power of the thing was unbelievable. The bottle literally vanished from the pad with a wet sort of whap, and when I looked up to find it, it was sailing over my neighbor's house, then their neighbor's house. Holy smokes! It finally landed in a driveway two doors down. When I recovered from the shock, I whooped for joy, raced out to the street to retrieve the bottle, and knew I was hooked. It's all been downhill from there.
Back to ContentsJust slapping a nose cone and some fins on a bottle may not get you what you want. Although a complete treatment of rocket aerodynamics is far beyond the scope of this page, here's a little info on the basics. For a more complete treatment, I highly recommend reading the appropriate chapters in the "Handbook of Model Rocketry" by G. Harry Stine. There are basically two problems to solve: making the rocket aerodynamically stable, and minimizing drag.
The first thing is to make the rocket aerodynamically stable. For a rocket to fly in a stable fashion, the center of gravity (CG) must be forward of the center of pressure (CP). The CG of the rocket is easy to find: it's the point at which the rocket balances. If you were to tie a string around the rocket at its CG, it would hang from the string horizontally. The CP is more difficult to determine. The CP is defined as the point along the rocket where, if you were to attach a pivot and then hold the rocket in the wind by that pivot, the wind forces on either side of the CP are equal, so the rocket wouldn't point either into or away from the wind: it would be "wind balanced." Adding fins to a rocket moves the CP toward the fins, since you're adding surface area. In fact, that's the main function of fins: moving the CP aft.
There are a couple of methods of determining the CP of a rocket. One old standby is to make a cardboard cutout shaped like the silhouette of the rocket, and then find the cutout's balance point, or CG. This corresponds pretty closely to the CP of the rocket itself. The other method is to calculate it, and the equations for doing so can be found in the "Handbook of Model Rocketry" mentioned above, as well as many other places. The equations are a bit hairy, but naturally there are several computer programs to help. Many, many rocket-oriented programs are available in the various "programs" directories of the model rocket ftp site. (By the way, there are several metric tons of information about rockets at this ftp site, it's well worth an extended browse.) One recommended program is VCP for Windows.
One question remains: how far forward of the CP should the CG be? Generally it's recommended that the distance between them be 1 to 2 times the diameter of the rocket body (this is called 1 to 2 caliber stability). This is just a rule of thumb, but it's a very good one.
(Note: most water rockets are designed to be stable when empty. However, the few I've seen that are stable even with water in them - usually the fins are large and hang down from the cap of the bottle - fly beautifully. Further, with a rocket like that you can make the nozzle smaller - and thus the thrust longer - without worrying about the rocket getting off course.)
The second thing is to reduce drag to a minimum. At the high velocities these rockets achieve, air drag becomes a very significant force. Again, see the "Handbook of Model Rocketry" for details, but in general the fins should be thin, every surface should be as smooth as possible, the nose should be a reasonable shape (it turns out that a hemisphere isn't bad, and is usually easy to come by), and the cross-sectional area of the rocket as small as possible. Making the rocket body from multiple narrow bottles is one way to achieve this (see Bottle Joining Methods below).
Back to ContentsThe simplest kind of rocket doesn't really require a launcher at all, just some way to prop the bottle in position while you pump. This kind of rocket is made by simply stuffing a rubber stopper in the neck of the bottle, inserting an inflation needle through a hole in the stopper, and pumping it up until it blows! You never quite know when it's goiing to go, which of course adds to the excitement (kids love this). There's lots of good information on this sort of launch technique at the Interplanetary Water Rocket Society home page, maintained by Gordon McDonough in Santa Fe, New Mexico.
The next step up is to build or buy a launcher that will hold the bottle down while you pump, and has some sort of trigger arrangement to let the bottle go. The POP-FLIGHT launcher mentioned above is the cheapest commercial launcher I've seen, and is quite adequate for initial forays into this hobby. It's design, however, limits you to using "full bore" nozzles, and doesn't allow for swept fins. There are a few other commercial launchers available, one from the 4H club, a couple from scientific or educational suppliers, and the Bigfoot launcher described below, which is in a class by itself.
If you like building things, a decent launcher is fairly quick and cheap to build. There are lots of launcher designs that work well. The simplest and easiest to build launcher I've seen yet is on Ian's Water Rocket page, it uses cable ties for a release mechanism, brilliant! Another cheap and easy to build launcher was designed by a retired machinist I know. Bruce Berggren in Texas has an excellent water rocket site that includes some essential threoretical info, some rocket construction tips, and a diagram of another excellent launcher, built primarily from PVC pipe. Kamiel Martinet's site also has detailed instructions for a great launcher, and excellent illustrations to go with them.
A launcher can also be built from a plastic quick-connect hose coupling, by epoxying the nipple part of the connector into a drilled out bottle cap, and attaching a tire valve to the "barrel" part. The flight of the rocket is different with this type, since the nozzle is much narrower than a plain bottle neck. This seems to be the predominant method in Japan, according to our Tokyo correspondent Shushi Shiota, who has some pictures of such a launcher at his site in Japan. Another quick-connect launcher, this one hand-held, is detailed on Clifford Heath's site. And yet another is on my long tailed rocket page.
Another type of launcher is described in the article "It's Launchtime!" in Science and Children magazine, Feb. 1993. And those plans were adapted from plans in "The Rocket Project", Science Scope magazine 15(2), 1991. I haven't seen the latter, but the former was the basis for two reasomably workable launchers I've built, one for some friends in Arizona, the other for my friend Ned in Seattle (a picture of the latter is here). However, having seen lots of launchers, I'd recommend one of the others over this "scissor" type.
Finally, if you want to spend some serious money, the Cadillac - no, the Rolls Royce - of water rocket launchers is the Bigfoot launcher manufactured by Gary Ensmenger. This launcher is truly deluxe, with an excellent, patent-pending hold-down and trigger mechanism, safety pressure relief valves, a built-in pressure gauge, a hose connection so you can fill the rocket on the pad, and many other great features. Check out Gary's excellent site for more info on his launcher, tips on rocket construction, ideas for experiments, and lots more.
Back to ContentsI don't have too much to say about pumps, except that this application is punishing: you need a good, heavy duty pump if you plan to repeatedly pump your rockets to 100 psi or higher. The usual wimpy bike pump just can't take it. Gary Ensmenger's site has more information about this, and he sells a great pump for a very reasonable price.
Back to ContentsThere are 2 methods for connecting bottles together in an airtight way that are currently in wide use. One is a purely mechanical (no glue, easy to assemble/disassemble "in the field") method we've been calling the "Robinson Coupling", which is fast and easy to make. The downside is that it makes a very narrow connection between the bottles, which has some disadvantages. The other method is a permanent, bonded circumferential splice, detailed on Bruce Berggren's page of construction tips. This one is really better, performance wise, but requires more care and time, and is a permanent join.
Back to ContentsI developed a fairly reliable and easy to build parachute release some time ago. It's velocity sensitive, triggering release when the rocket slows down at apogee. This only works for near-vertical launches, of course. The idea is simply to have flap hinged to the body of the rocket and biased so that it wants to swing away from the rocket body. Air streaming past the rocket holds the flap down as long as the rocket is moving fairly fast. When the rocket slows sufficiently, the flap opens, releasing the nose cone and the parachute inside. A variation of this release that uses no metal has been developed for use in the Science Olympiad events, which prohibit metal parts.
This release works well, but is a bit "finicky", and work is ongoing to develop a better one. An electronic timer is promising, though wouldn't be allowed in any "non-metal" rocket events. Bruce Berggren has had great success using the clockwork mechanism from wind-up toys as a timer, this works very well. See his site for details.
There have been some experiments with helicopter recovery as well, and these are ongoing. The same hinged flap arrangement shows promise, using two or more flaps as the helicopter blades.
Of course, the simplest recovery system is none at all, and this is fun, if a bit hard on the rockets. My oldest surviving rocket, the HOV-1 (HOV stands for Hoary Old Veteran), has a duct tape and rubber nose cone: the nose absorbs enough energy that the rocket repeatedly survives its very hard landings from 300 feet. In fact, this rocket bounces dramatically, sometimes up to 20 feet in the air, leading to an entirely new kind of recovery system: bounce recovery. Engineers are currently investigating the potential commercial uses of this system.
Back to ContentsBruce Berggren came up with a wonderfully easy to build pressure sensitive staging mechanism, based on ideas from Adrian Righetti in Australia. However, Adrian's device requires some custom machining. Bruce's mechanism is built from commonly available, off-the-shelf parts. Check it out.
There are 2-stage toy water rockets available, and I purchased one of these to see how it worked. Unfortunately, it didn't. (The one I bought was a cheap knock-off of the "real thing" and manufacturing defects ensured that it wouldn't function. Planned obsolescence? You be the judge.) However, I could see how it wassupposed to work, and the mechanism may be useful for our larger, more powerful rockets, though will probably require some custom machining of parts.
Back to ContentsHere's a brief trip report from my recent trip to Thailand and Japan. It was mostly a normal vacation, but rocketry crept in a bit, especially in Japan, where water rockets are very popular (in fact the main reason I went to Japan was to meet with other water rocketeers - the First International Water Rocket Symposium!).
Here's a couple pages with photos and captions from the "Idea" water rocket contest in Japan in the summer of 1997. Need an idea? There are plenty of them here...
There are about a bazillion sites dealing with water rockets out there, try typing "water rocket" into your favorite search engine if you have a few hours to kill. Here are a few to get you started:
There's now a mailing list for water rocket folks. To sign up send mail to majordomo@lists.osa.com.au with the message "subscribe water-rockets" (without the quotes) in the body of the message. Expect several messages a day.
There's also a downloadable zipped text file (693 K) containing all the rocket mail I've saved since 12/96, all 1450 messages!! Many of them precede the water rocket mailing list. It's 2.5 Mb of text when unzipped. Naturally there's a lot of junk in there, but for those with the patience to wade through there's a wealth of information to be gleaned from it.
Back to Contents Back to Top