Re: How can you measure how fast sound travels?

Hi!

It seems I'm getting all the "speed of sound" questions. Below you find 
two links to the MadSci archive describing one particular method of 
measuring speed of sound in air, the "Kundt tube":
 http://www.madsci.org/posts/983199157.Ph.r.html http://www.madsci.org/posts/archives/jul2000/964817012.Ph.r.html

For science classes this experiment is the easiest to perform and the most 
instructive. You will find that the Kundt tube experiment has two 
disadvantages namely that it is only practical in air and that it measures 
the speed of sound inside a tube, which is not necessarily the same as the 
speed of sound in free air (though the difference is admittedly small).

The latter is actually a basic problem in determining the speed of sound 
in general. It is not only dependent upon the medium (and the state it is 
in) but also on the shape of the medium. This is particularly tricky when 
the wavelength is of the same order as the smallest dimension of the 
medium. Conversely, the speed of sound is wavelength dependent.

Something like the kundt tube experiment can be used to determine the 
speed of transversal waves in solids. Such waves are found on metal plates 
which are made to resonate. For example, ever tried playing a cymbal using 
a violin bow? If you put some dry sand on the cymbal the standing waves 
generating the shrieking sounds will be visualised by the sand settling 
down on the nodes.

For longitudinal waves in liquids and solids it is not easy to visualise 
standing wave patterns. Some indirect way of detection must be used 
instead. A bar of iron can be struck with a hammer on one end and the 
ringing requency measured. For liquids I can't really think of an easy way.

Probably the most generally applicable method of measuring the speed of 
sound is the direct method: apply a pulse (a very short burst of energy) 
and measure at some distance of the source how much time it took the 
impulse to make its way there. To eliminate measurement errors caused by 
the unknown delay in the source you might use two identical sound 
transducers (for gases: microphones, for liquids: hydrophones, for solids: 
accelerometers) spaced a known distance apart and measure the time 
difference between those two. It be said that such a setup is quite 
specialised and expensive.

Conclusion: Practical experiments not involving precision measurement 
equipment rely on creating and detecting standing waves in the medium to 
be measured. From there, the speed of sound can be calculated.

Regards,

Bruno