Lecture 9: Normal Modes in Sound and Music

OCW Scholar

« Previous | Next »

Lecture Topics

White sand forms figures that look like hourglasses on a black piece of paper.
  • Musical Instruments
  • Sound Cavities
  • Normal Modes

Learning Objectives

By the end of this lecture, you should:

  • understand how normal modes are controlled by boundary conditions.
  • describe the basic structure of a stringed instrument.
  • explain the role of tension, mass per unit length, length, and boundary conditions, and/or speed of sound, as appropriate, in producing notes.
  • write the wave equation in higher dimension than 1, in Cartesian coordinates, and solve, incorporating boundary conditions.
  • explain the principle behind Chladni patterns.

Lecture Activities

Check Yourself

  • In the lecture, Prof. Lewin states that if one end of a string was free (hard to do and still have it move in one plane), the boundary condition would change to having the slope at the free end be zero. Explain why this is.

View/hide answer

A tension in the string is required to have simple wave propagation such as we discuss (in fact we assume the tension is the same at all parts of the string). Whatever "ring" might be used to produce this tension would not only have to be friction-free (or else the end would not be "free") but would have to be of very low mass if not to load the string. In that case, if there was any slope in the string, the tension in the string would apply a transverse component to the ring, which being massless would have infinite acceleration. The only way this can be avoided is if we use the boundary condition that the slope is zero. Another way to look at this is that the massless ring can move very fast and always cancel out any force that would arise if the slope was not zero, and in that way maintain the slope at zero.

 

  • Explain why an open end in a wind instrument is a node, amplifying slightly the argument made in the lectures.

View/hide answer

Sound waves arise from overpressure (in fact very slight increases in pressure compared to ambient atmospheric pressure). In the vicinity of an opening and in the absence of unusual conditions like supersonic shock waves, pressure equalizes very rapidly. Thus there can be no overpressure near an opening, and the overpressure variable which represents the sound wave must be zero there, making a node.

 

  • Water waves are in general rather complicated, and despite being the most familiar sort of wave, are not studied in detail in this course. However, a thought experiment about the familiar situation of water sloshing in a bathtub is interesting! Consider a bathtub of 1.5 m by 1 m, with vertical walls, and that the speed of waves is \(v = \sqrt {gh} = 3.1\sqrt h \), where h is the depth of the water. This equation is for "shallow water" compared to the wavelength, and we expect the wavelength to be about the size of the tub, so let us take h = 10 cm, to give a speed of \(v = 3.1\sqrt {0.1} = 1\)m/s. What standing wave might you get going by moving your hand up and down in the middle of the tub?

View/hide answer

This took a bit of work to describe, but the solution is fairly easy. There will be antinodes at the walls and in the middle where you are moving your hand. The waves will have to be in the second harmonic in both directions (there are two nodes inside the tub in each direction). If the long direction is in x and one corner the origin, we get a water surface displacement \[z(x,y,t) = {A_{2,2}}\cos \left( {\frac{{2\pi x}}{{1.5}}} \right)\cos \left( {\frac{{2\pi y}}{1}} \right)\cos ({\omega _{2,2}}t)\]where \[{\omega _{2,2}} = v\sqrt {{{\left( {\frac{{2\pi }}{{1.5}}} \right)}^2} + {{\left( {\frac{{2\pi }}{1}} \right)}^2}} = 1(7.55) = 7.6rad/s\] or 1.2 Hz. Note: this actually worked quite well in the course developer’s tub! It is a slightly unusual Japanese-style tub with vertical sides and quite rectangular: results may be more complicated in other tubs.

 

  • If you go into a completely empty room with smooth hard walls and a closed door, you will find it quite "echo-y". Consider yourself to be at the exact center of a room 3 m by 4 m and 2 m high. Write the solution for waves you could excite by making a noise there, and determine what the lowest mode you could excite would be. Recall that the speed of sound is 340 m/s.

View/hide answer

The general solution would be \[{p_{l,m,n}} = A\cos \left( {\frac{{l\pi x}}{4}} \right)\cos \left( {\frac{{m\pi y}}{3}} \right)\cos \left( {\frac{{n\pi z}}{2}} \right)\cos {\omega _{l,m,n}}t\] if the longest direction is in x, z is vertical, and the origin is on the floor at a corner. The angular frequency is \[{\omega _{l,m,n}} = v\sqrt {{{\left( {\frac{{l\pi }}{4}} \right)}^2} + {{\left( {\frac{{m\pi }}{3}} \right)}^2} + {{\left( {\frac{{n\pi }}{2}} \right)}^2}} \] By making a noise in the center there would have to be an antinode there, much as at the walls, so in all directions there would be second harmonic. So the actual lowest mode solution is \[{p_{2,2,2}} = A\cos \left( {\frac{{2\pi x}}{4}} \right)\cos \left( {\frac{{2\pi y}}{3}} \right)\cos \left( {\frac{{2\pi z}}{2}} \right)\cos {\omega _{2,2,2}}t\] with angular frequency\[{\omega _{2,2,2}} = 340\sqrt {{{\left( {\frac{{2\pi }}{4}} \right)}^2} + {{\left( {\frac{{2\pi }}{3}} \right)}^2} + {{\left( {\frac{{2\pi }}{2}} \right)}^2}} = 1390 rad/s. \] This is a frequency of 221 Hz, a common frequency in speech and near middle A on a piano. This explains why the room would sound "echo-y": part of that effect would be resonance. With a sound generator you would likely be able to explore many node effects in such a room!

 

« Previous | Next »