Sunday, February 26, 2012

PVC Steam Whistle

•1/2''-1'' pvc pipe
•1/2''-1'' wood dowel

• fine toothed wood saw and a miter box or a scroll saw
• a chisel and hammer

The first step is to make a flat on the lower dowel plug.  You can do this either by cutting into the end of a long dowel with a saw (at about 1/3 the diameter) then cutting off the two slices about 1'' long,  or cutting off a section of dowel about 1'' long and slitting it with a chisel and hammer to create the flat.

The next step is to cut out the mouth of the whistle, cutting a little less than half way into the diameter and then in at a 30 to 45 degree angle to remove the slice.

Insert the dowel with the slice taken out of it into the lower side of the whistle so the flat faces the mouth of the whistle. The top of the dowel should be level with the bottom lip of the mouth for the best sound.   Note that pvc and dowels are not all that consistent in diameter so you might have to add some masking tape to the rounded side of the plug to keep it from falling out.

Cut your pipe to a little longer than the desired length measuring from the top of the bottom plug and add the upper dowel to complete the whistle. If you want to be really precise the length of the plug in the tube should equal the extra length you added to the whistle pipe.  

Playing the the whistle
The specific tone of the whistle is determined by the interior height of the tube (the length from the top to the bottom plug).  While the whistle's length determines the natural frequency it wants to play, over blowing the whistle can cause the tone to shift upwards to a quieter breathy sounding overtone. A simple a approximation for tone is F= V/4L, the frequency (F) will equal the speed of sound (V) over 4 times the length of your whistle tube (L), as closed tubes tend to resonate with a quarter wave length.
To simplify things we tend to reference the length/pitch table used for palm pipes. For the full set of equations see this wikipedia article on acoustic resonance. Note that if you get much over 8'' you'll have to switch from 1/2'' pvc to 3/4 or 1'' to prevent immediate overblowing.    

Useful links
A good overview of whistle engineering can be found at this site. The author provides several optimal ratios (length to diameter and mouth size to whistle diameter)  for making a loud and consistent whistle. Playing with our own whistles we found his numbers to be a bit conservative, though the whistles with a length to diameter ration of 3:1 are loud and do not over blow easily, you can stretch this ratio to about 16:1 for low notes with small pipes and get away with it.  The drawback is of course they can over blow more easily.  

Here's a Multi tone (chime) steam whistle cleverly put together using pvc elbows and tee fittings.

Note a variable pitch slide whistle version will be coming soon.

Monday, February 20, 2012


•1/2'' dowel (about 2' long)
• coffee can (we like the kind with steal bottoms and cardboard sides as its easier to put holes in them)
• 50lb test fishing line
• small eye screw
• a finishing nail
• a cork or a small piece of wood

• scissors
• hammer 
• a file or a small saw 

To start the instrument carefully use a nail to mark the position for the dowel hole, making sure that its low enough that the dowel will not touch the surface of the can when inserted.  When you're satisfied with the positioning of the holes use a pair of scissors to widen them out so they are just smaller than your dowel. 

In the past when we were making these with large groups we would pre-prepare the cans to this step, as there is tremendous learning potential in these steps and mistakes here can hinder the function of the instrument later.

Sand the corners off of your dowel so it can be press fit into the can wall as seen above.

Hammer one nail into the dowel about 1/2'' from the end and use another nail to make a pilot holes about 1/2'' to an inch from the other end of your dowels making sure that they both face up. Its a good idea to bend your nail toward the end of the dowel (unlike the picture) to prevent the string from flipping off and from making a hazardous spike.  Note that you can do this part before or after you add the can to your dowel, hammering with the can on the dowel can be awkward but chances are it won't hurt the instrument.

Take the eye screw and screw it in 3 or 4 turns into the pilot hole. Note that the dowel must be in the can at this point as you won't be able to get it afterward.

Tie your fishing line around the nail and bottom of the dowel to secure it.  Again make sure to bend the nail down.  Make sure the eye screw is mostly unscrewed and tie the fishing line onto it. Then screw it in to tighten the line by twisting it around the shaft of the screw.

Once the the line is taut you can play the canjo but it will produce only a soft dull sound. To rectify this we need to transfer the energy of the vibrating string, which has very little capacity to move the air around it, to the head of the can with its much larger surface area thus air moving capacity. To do this we need to make a bridge, a small piece of wood or cork with a groove filed into it for the string to rest in. This can be achieved with the corner of a flat file or a small saw.

The end result being something like this,

or this (which can be done with a utility knife)

To finish add the bridge between string and the top of the can.

After the canjo is assembled lots of experiments can be done to figure out how to change the tone (frequency of the sound), timbre (the character of the sound) and volume.  Tightening and loosening the string or fingering like a guitar are some ways to change the tone, tighter and shorter makes high notes and looser and longer makes low notes. Moving the bridge, closing the can and adding holes, cutting the can off, or adding other items to the can are ways to change the timbre of the sound and the volume.  You can add frets by wrapping wire around the dowel at specifics lengths. This will allow one to play written music, skill and patience pending, we've actually had several kids figure out how to play base lines of songs them knew on a canjo.  Being so simple there are a lot of ways to experiment with it to significantly change the sound it produces.  That said the form seen above is really a starting point, figuring out how to play it and change the sound it makes is the real value in this project.

Here's the canjo in action:

Sound Projects

There are a lot of good sound projects out there, and in the past when we've explored sound with students we've started with a variety of wind instruments developed by other organizations.  Below I will list some some of our favorites, with some modifications and technical notes.

Slinkys and waves:
At the most basic conceptual level sound is about waves, but not like ocean waves that oscillate up a down but rather waves of compressed material traveling through space. Sound, to many kids surprise, can travel  through any type of matter, traveling faster in non-compressible materials like water and metals and slower compressible things like gasses and spongy solids. To complicate matters further sound will travel faster in light material than in more massive material. The best simple demonstration of this we found is The Exploratorium's Slinky in Hand  science snack, it gets at the difference between a transverse wave (a water wave b.) and a compression wave (a sound wave a.) and begins to discuss the idea of resonance. They suggest that you use monofilament to demonstrate a compression wave, while this may make everything a little clearer we never found it unnecessary as you can get good results just by stretching the slinky between 2 people and then having one person hit the back of the hand holding the slinky with their free hand to make a compression wave.

Palm pipes:
Wind instruments are a good way to get at the concept of resonance. To correlate the idea that it's mostly the length of air enclosed in the instrument that controls the pitch of the sound it makes. Somewhere between the slinky demonstration and wind instruments are tuned percussive instruments, think of it as a single pulse wind instrument.  Palm Pipes are a simple way to connect the length of the resonator, in this case a piece of PVC pipe, to a specific frequency sound produced.  If you search Palm pipes you'll find lot and lots of different sites, but here are a few we think is fairly thorough, here and here. By blowing over the pipes (try open and closed) or hitting them on your palm as the name suggests each length will produce a specific note.  By giving each student a pipe section and asking them to figure out how to get their PVC pipe sections to make noise (find ways to vibrate the air inside them) and then to identify what physical characteristics determine tone vs timbre is a useful exercise.      

Straw Oboe:
After Plam Pipes, the Straw Oboe is probably the simplest known wind instrument consisting of a straw that has been flattened at one end and then had the corners cut off to make a oboe like reed.  You can demonstrate that the length controls the pitch (the frequency of the sound waves) by playing the straw while cutting off the end with scissors, the pitch will get higher and higher as the straw get shorter.   For a less destructive way of adjusting the pitch you can add finger holes to the straw. While the Science Snack suggests using a soldering iron we found that just folding the straw and cutting off the corner of the fold produces good finger holes.  

If a finger hole is uncovered the sound wave treats it as if the straw ends at that point bouncing back early allowing you to change the pitch dramatically.

Despite is extreme simplicity Straw Oboes can be hard to use for younger kids as you may need to squeeze the straws a little with your lips to get it to play. This is especially true if you have really larger stiffer straws (jumbo straws), so go for the smaller squishy ones (not the jumbo straw in the picture).  

While the Glove-a-Phone (from RAFT) has more parts it can be easier to use than a Straw Oboe and makes a much deeper and somewhat less obnoxious sound. We have made a few simple improvements to the original design. We tend to use a Nitrile instead of rubber gloves as they are hypo allergenic and we found they are less prone to breaking from the vibration. We also found that the straw can come out quite easily and is very hard to put back in. By rolling the glove around the straw and then taping way past the connection and spiraling up the straw, it is much more securely attached and fewer gloves are wasted.  

 Another thing we spent some time on was figuring out how to allow the Glove-a-Phone to play notes outside of range determined by the length of the tube and the tightness of the glove.  We tried making finger holes but found that they had to be really big to change the pitch significantly, we suspect this has to do with the ratio between the diameter of the tube and finger hole size being too high.  However making a slide out of construction paper (a trombone Glove-a-Phone) works very well allowing major changes in tone not possible by just stretching the glove.     

In both the case of the Straw Oboe and the Glove-a-phone you have a valve (the reeds of the straw or the membrane of the Glove-a-Phone) and a resonator (the straw or cardboard tube). When you blow into the instrument the valve releases a pulse of air and then closes because of the vacuum in the straw or tube after the initial pulse of air.  This pulse of air travels down the tube with only vacuum behind it and when it exits the tube air from outside the tube rushes in to fill the vacuum, this second wave travels back to the valve and collides with it pushing it opening and allowing the next pulse of air to exit the valve.  Given the speed of sound in air is more or less constant, the biggest factor in the frequency of the pulses is the length the wave has to travel from the valve to the end of the tube and back.

Its possible to explain resonance in wind instruments in the form of a relay like game.  You need three people. One person to represent the valve at the top of your wind instrument, one to represent the bottom of the tube (the clapper), and one to represent the sound wave traveling between them (the runner).  To simulate the resonator in your wind instrument the valve person tags the wave who runs to the end of the tube tags them and then runs back to the valve to get tagged and repeats the process.  Every time the runner get the the end of the tube that person claps, by changing the distance for the runner to travel from 20 feet down to zero the clapper will clapper more and more often.  If you wanted to make it a true relay you could replace the valve person with a cone and have each returning runner tag a new runner. This would be closer to reality as new air is emitted with each pulse.

Length is not the only factor in changing the pitch, in the Straw Oboe the size of the straw and how close the reed flaps are can change the tone produced slightly as this changes how fast the reed is able to open and shut.  Biting down on the reeds softly tends to raise the pitch as the reeds can close more quickly. In the Glove-a-Phone using smaller tubes or tightening the membrane by pulling or pushing on it can also raise the pitch as it forces the membrane to open less and close faster.