Triple Singing Tubes

Demonstration Kit


Mysteriously play an organ pipe without an organ! This demonstration uses heated air to produce vibrations inside a long tube. The vibrations, in turn, produce standing sound waves with a unique tone, or timbre (tăm br)—the same concept that produces sound from an organ pipe.


  • Sound waves
  • Wind
  • Open-ended resonance tubes
  • Organ pipes


Bunsen burner, or portable laboratory burner
Heat-resistant gloves or oven mitts
Metal tubes, 3 (2" dia. x 13⅛", 19½", 19¾")*
Metal wire disks, 2¼" diameter, 9*
Paper clip, metal
Pliers, needle-nose, with wire cutters
*Materials included in kit

Safety Precautions

The edges of the metal wire disks are sharp. Please handle with care. Follow normal Bunsen burner safety guidelines. The metal tube and wire disks will get hot while in the burner flame. Wear heat-resistant gloves and safety glasses when performing this demonstration.


The materials are completely reusable and should be saved for future demonstrations.

Prelab Preparation

  1. Obtain the metal wire disks. Handle them very carefully.
  2. Carefully bend one of the disks into a bowl shape (similar to a watch glass shape). Put pressure on the center of the disk with your thumbs and then evenly bend the edges of the disk with your fingers to form a bowl (see Figure 1). Be very careful not to cut yourself on the wire edges.
  3. Form a bowl shape that has a slightly larger diameter than the inside diameter of the metal tube. If the wire disk is bent too much, it can be flattened out and adjusted to the proper diameter.
  4. Repeat steps 2 and 3 for the other wire disks.
  5. Obtain three bowl-shaped wire disks, and place one disk on top of the other to make a stack of three disks.
  6. Obtain a metal paper clip and needle-nose pliers with wire cutters.
  7. Straighten out the paper clip.
  8. With the wire cutters, clip off a 1–2 cm paper clip piece.
  9. Insert this paper clip piece through the center of the wire disk stack so that it goes through all three disks (see Figure 2).
  10. Use a needle-nose pliers to bend the inserted paper clip piece into a “C” shape to secure the wire disks together (see Figure 2).
  11. Insert the curved end of the wire disk “stack” into the end of one of the metal tubes, opposite to the label (see Figure 3). Carefully push and “massage” around the edges of the stack evenly in order to slide the disks into the tube so that they remain parallel with the tube opening. (When you look inside the tube, there should be no gaps between the disks’ edges and the wall of the tube.) The friction between the wall of the tube and the edges of the disks should keep them secure inside the tube. If the disks are loose or fall out, remove them and flatten them out slightly to increase their diameter so they fit snugly inside the tube. If the wire disks are pushed in crookedly, use pliers to remove them completely and begin again. Proceed slowly in order to keep the wire disks parallel with the opening of the tube.
  12. Once the wire disks are inside the tube opening squarely, slide them down the tube so that they are about 1–2" from the end of the tube (see Figure 3).
  13. Repeat steps 5–12 for the other wire disks and metal tubes.
  14. Set up a Bunsen burner or portable laboratory burner on a demonstration table.


  1. Light the Bunsen burner and adjust it to obtain a blue flame.
  2. Wearing heat-resistant gloves or oven mitts, position the 19¾" long, or 19½" long metal tube vertically over the Bunsen burner flame to directly heat the wire disks inside the tube (see Figure 4)
  3. Heat the wire disks for 10–15 seconds. Swirl the tube slightly to evenly heat the entire disk surface.
  4. Remove the tube from the Bunsen burner flame and continue to hold it vertically. Notice the loud tone that reverberates from the tube. The sound will last for several seconds (10–30 sec.), depending on how quickly the wire disks cool.
  5. When the sound disappears, repeat steps 16–18 as often as necessary to reheat the wire disks and reproduce the sound.
  6. After the initial demonstration, perform this variation: After heating the wire disks, remove the tube from the burner flame and immediately tip the tube 90° so that it is parallel to the floor. Notice that no sound resonates from the tube!
  7. Quickly, but steadily, rotate the tube to the vertical position. The sound gradually increases in volume as the tube rotates towards the vertical position. (This should be done in 5–10 seconds—before the wire disks cool off.)
  8. Repeat steps 16–21 with the 13" long tube. Notice the higher pitch that the shorter tube produces. (It will take longer for the shorter tube to resonate because higher frequency vibrations are required and need more time to develop.)
  9. Use the 19¾" and 19¾" long tubes together to produce a beat frequency: First, play the 19¾" and 19½" long tubes separately by following steps 16–18. Notice that both tubes produce a similar pitch.
  10. Then, hold the 19¾" tube in one hand and the 19½" tube in the other so they are next to each other and vertical.
  11. Heat the 19¾" tube in the burner flame for 10–15 seconds following steps 16–18.
  12. Remove the 19¾" tube from the flame and then quickly heat the 19½" tube for 5–10 seconds following steps 16–18. (The 19¾" tube will begin to “sing.”)
  13. Remove the 19½" tube from the flame and place it near the “singing” 19¾" tube. The 19½" tube will begin to “sing.” Notice the pulsating tone as the sound volume varies between loud and soft at regular intervals. This is a beat frequency. The beat frequency demonstrates that the tubes are resonating at slightly different frequencies.
  14. Discuss the observations with your students.

Teacher Tips

  • The materials in this demonstration kit can be reused many times. If the metal wire disks deteriorate, or are lost, replacement metal wire disks are available through Flinn Scientific, Inc. (Catalog Number AP6313; package of 10)
  • Three wire disks are inserted into the tube to provide a large surface area to retain heat, and to create very tiny holes which produce more turbulent airflow. This generates a Singing Tube that “sings” for a longer period of time (20–30 seconds). Two wire disks can also be used to shorten the time that the tube will resonate. A tube with one wire disk will only “sing” for a few seconds.
  • Heating the wire disks with a double-cone Bunsen burner flame will heat them quickly and to a very high temperature. This will produce a longer-playing Singing Tube. However, the high temperature will also quickly deteriorate and weaken the wire disks. Do not heat the wire disks with a double-cone flame for more than 10 seconds. Allow the disks to cool for a longer time in between trials when using a double-cone flame.
  • Two Bunsen burner setups can be used to heat the 19¾" and 19½" tubes simultaneously when demonstrating the beat frequency.
  • Do not allow the metal tubes to get too hot. The heat may scorch the tube label, heat-resistant gloves, and/or lab table they are stored on between trials.
  • The Singing Tubes can be silenced quickly, if they are too loud, by turning them so they are horizontal, or by covering one of the ends (preferably the cool end).
  • This is a great demonstration to begin a sound unit, as well as when discussing heat transfer and wind.
  • You can pretend to pour sound into a beaker, or from a beaker into the tube. When the Singing Tube is “singing,” begin to pour the “contents” of the tube into a beaker by rotating the tube 90°, with a beaker at the top opening of the tube. As the tube rotates, the “singing” decreases and makes it appear as if you are emptying the sound from the tube into a beaker. Reverse the process by pretending to pour sound into the tube. Start with a heated tube held horizontally and then quickly, but steadily, rotate it to the vertical position as you hold the lip of a beaker near the “top” of the rotating tube. The volume of the “singing” increases, simulating the act of pouring sound into the tube.
  • A sound meter, or microphone, can be used to measure and visualize the harmonic frequencies produced by the resonating tubes. Determine the speed of sound in the tube by comparing the measured harmonic frequencies to the length of the tube.

Correlation to Next Generation Science Standards (NGSS)

Science & Engineering Practices

Using mathematics and computational thinking

Disciplinary Core Ideas

MS-PS4.A: Wave Properties
HS-PS4.A: Wave Properties

Crosscutting Concepts

Energy and matter

Performance Expectations

MS-ETS1-2: Evaluate competing design solutions using a systematic process to determine how well they meet the criteria and constraints of the problem.


Many musical instruments work because air is vibrated inside a column. The length of the column determines the sound produced by the vibrating air. The vibrations result in a mixture of different frequencies that may resonate inside the column if the frequencies match the harmonics of the column. The combination of resonating frequencies will result in a tone, or timbre, that will be unique to the column.

The open tubes in this demonstration act in a similar manner to the flue-type organ pipe (see Figure 5). In a flue-type organ pipe, a stream of air is directed against a sharp edge in an opening of the pipe. The sharp edge creates turbulent, complicated swirls of air which set up vibrations in the air column. The vibrations that are at the correct resonance frequency of the pipe (depending on the length of the pipe, the design of the pipe, and the air temperature) will resonate and produce a very loud tone. The tone is not a specific fundamental frequency, but it is a combination of the different harmonics that the column will allow. The fundamental frequency is usually the most prominent frequency in a resonating column. The shorter the pipe, the higher the vibrational frequency must be to produce resonation inside the column. Therefore, a short column will produce a higher pitch than a long column. (See Resonance and Harmonics Background for more information.)


The Singing Tube is an example of an open-ended resonating air column. When the metal wire disks are heated, and then removed from the heat source, the metal will retain the heat for a time. This heated metal will heat the nearby surrounding air, which then rises through the tube. As the hot air rises, cooler air from the room will flow into the tube from the bottom and through the wire mesh. When the air flows through the wire mesh it becomes turbulent. The swirling turbulent air sets up vibrations inside the tube, and the correct vibrational frequencies will begin to resonate loudly inside the tube to produce a note, just as in the organ pipe. When the tube is tilted parallel to the ground, the heated air does not rise through the column to cause a large inflow of cooler air through the wire mesh. Without the rush of cool air through the tiny holes, no vibrations, and therefore no sounds, are produced.

The beat frequency that is heard when the two nearly identical-length tubes “sing” together is a result of interfering sound waves of nearly identical frequencies. Sound frequencies that are slightly different will interfere with each other constructively and destructively to produce a pulsating sound-wave pattern. The volume of the tone varies between loud and soft in a regular interval, or frequency. (Complete destructive interference only occurs if the amplitudes of the interfering frequencies are the same.) The frequency of the pulsating tone is actually the difference in frequency between the two interfering sound frequencies. For example, if a 600-hz and 597-hz tone resonated near each other, a pulsating sound with a frequency of three hertz would be heard.

Resonance and Harmonics Background

All objects have a certain frequency or a set of frequencies at which they most easily vibrate. This is known as an object’s “natural frequency” or “harmonic frequency.” When a forced vibration on an object matches the object’s natural frequency, an increase in vibration will occur. When an object is forced to vibrate at its natural frequency, a standing wave is formed within the object.

A “standing wave” is a wave pattern that appears to be standing still. To create a standing wave pattern, two waves must constructively and destructively interfere. The two interfering waves must be traveling in opposite directions, and have the same frequency, wavelength, and amplitude. This commonly occurs when a vibration produced by a source is reflected off a medium. The reflected wave will then interfere with an incident wave created by the same source. It is important to realize that a standing wave pattern is not an actual wave, but rather a pattern created by the interference of two waves. Because of this, standing waves do not have the typical crests and troughs as other waves do, but rather nodes and anti-nodes. A node is a point on a standing wave that appears to be standing still due to complete destructive interference. An anti-node is a point on a standing wave halfway between two nodes, at which the largest amplitude occurs. Figure 6 represents a common depiction of a standing wave pattern.


Standing wave patterns are often set up in musical instruments that are plucked, or bowed. They are also set up in wind instruments by the vibrations of a reed or musicians’ lips. Standing wave patterns are only created at an instruments’ natural frequencies, also known as harmonic frequencies. The harmonic series for a string instrument and an open-ended air column are shown in Figure 7. Notice that for a string instrument, a node is present at the start and end of the standing wave pattern. This is because the ends of the strings are fixed and not allowed to vibrate. This is not true for wind instruments containing an open-ended air column. In an open-ended air column, the sound wave traveling through the tube is reflected back on itself by the air molecules outside of the tube. When the reflected wave interferes with an incident wave, an anti-node is present at the start and end of the standing wave pattern.


Look again at Figure 7 which represents the harmonic series. Notice the formulas written for each harmonic. These formulas were derived from the common formula below, which represents how the speed (v), wavelength (λ), and frequency (ƒ) of a wave are related. The formulas can be used to calculate the frequency of each harmonic, as long as the speed of sound and the wavelength are known.

v = ƒλ or ƒ = v/λ

Further analysis of Figure 7 shows that in order to calculate the wavelength (λ) for each harmonic of a string instrument, the length (L) of the string and number of waves must be known. The first harmonic shows ½ of a wave present within the string, therefore the string length must be doubled to get one complete wavelength. The second harmonic shows one complete wave within the string, therefore the string length and wavelength are equal. The harmonic pattern continues, as seen in Figures 7 and 8. By knowing the instrument’s string length and the number of waves, the actual wavelength can be calculated for each harmonic.


The formulas in Figure 7 can also be used for calculations involving the open-ended air column. The only difference here is that you would need to know the column length (L) for each harmonic instead of the string length. Although the standing wave patterns of a string instrument look different than an open-ended air column, the number of waves in each harmonic is the same. For example, the 1st harmonic for the opened-ended column shows ½ of a wave within the air column (this may be hard to see). The air column length must be doubled to get one complete wavelength. This is exactly what was done for the string instrument in the first harmonic. Further analysis of the diagrams in Figure 7 reveal that the number of waves present in each harmonic is the same for both the string instrument and open-ended air column.


Flinn Scientific would like to thank David Katz, Pima Community College, Tucson, Arizona, for providing us the idea for this demonstration.

Tipler, Paul A. Physics for Scientists and Engineers, 3rd Ed., Vol. 1; Worth Publishers: New York, 1990; pp 452–457.

Next Generation Science Standards and NGSS are registered trademarks of Achieve. Neither Achieve nor the lead states and partners that developed the Next Generation Science Standards were involved in the production of this product, and do not endorse it.