Skip to the content

Websites we've worked on:
Sound tourism map link
sound around you link
sound101 link
Dr Andy Moorhouse on piano ...

BBC radio docs on sound:

- presented by Prof Trevor Cox

Breaking glasses using sound, continued

Second experiment

For first glass breaking page, click here, it is recommended you read this page first.

In the second experiment, we tried to increase the sound level produced so we could break glass without faults.

Experimental set up

We placed a loudspeaker at the end of a clear tube (see photo to the right). The loudspeaker was a 10" driver rated to 200W (sensitivity unknown). A piece of (white) wood was placed half a wavelength away so that the inside of the enclosure resonated at the resonant frequency of the glass. This enclosure resonance provided additional sound energy (when compared to an anechoic termination, even allowing for the loading on the loudspeaker). By placing the system in the tube, energy is not wasted due to spreading. The procedure is described below.

tuning the resonance

First, the generator feeding the loudspeaker was tuned so it produced a pure tone at the resonant frequency of the glass (this was described on the previous page). Tuning is done at a moderate sound level.

Then the glass was placed in the tube near the loudspeaker, but not touching it. It was found that the maximum pressure was achieved nearest the loudspeaker rather than the wood termination (this was measured on the microphone shown at the bottom of the first picture). The piece of wood was moved up and down the tube until the tube resonant frequency matched the glass' resonant frequency. This was found by finding the position of the wood that gave the maximum level measured at the microphone.

The output of the loudspeaker was then increased. The highest measured sound level we produced was 158 dB. Hearing protection is needed by the operator.

Broken glass

The glass then breaks, or rather it doesn't. We tried lots of different glasses, and none of them broke unless we introduced a small fault on the glass (for example a small scratch). This scratch doesn't have to be visible, but without it, the glasses were too perfect and there was no fault along which a fracture could grow. You can scratch the glass using a tile cutter.

The glass doesn't break easily, because although the sound level is very high, the excitation of the glass is very inefficient. At the frequencies we were generating, the wavelength of sound is very much larger than the diameter of the glass. Opposite sides of the glass need to move in opposite directions to create the correct resonant mode for the glass to break. But this is difficult to achieve with an airborne wave because the pressure on opposite sides isn't that different because of the large wavelength. Hence, while we produced very high airborne sound levels, the vibration introduced into the glass was relatively small.

But we now have a more reliable method: A more reliable way of breaking glass

Email the team

Re-use of material permitted provided it is clearly labelled with an appropriate credit such as: "(c) University of Salford,"