Natural Science Forum / Physics / Acoustics / February 2005

I can not explain it in simple terms.  Perhaps someone else can.  

In a constant diameter, infinitely long tube, dynamic pressure and dynamic
particle velocity are related to each other by a constant of
proportionality called the characteristic impedance of the tube.  Acoustic
impedance is defined as the ratio of pressure to volume velocity.  A cap on
the end of the tube represents a condition of high impedance and imposes
the requirement that pressure be high and that particle velocity be low at
the plane of the cap.  Similarly, an open end represents a condition of low
impedance and imposes the requirement that pressure be low and that
particle velocity be high at the plane of the opening.  Both of these
conditions impose an abrupt change in the relationship between pressure and
partical velocity as the wave travels down the tube.  It is the abrupt
change in the relationsip between pressure and volume velocity that creates
the wave reflection, not your so-called "thin air."

The situation of a sound wave traveling in a lossless tube is
mathematically identical to electromagnetic wave traveling in a lossless
transmission line.  The capped tube is analogous to an very high impedance
at the end of the electrical transmission line, and an open tube is
analogous to a very low impedance at the end of the electrical transmission
line.  In both cases wave reflection will occurr.  The only case in which
wave reflection will not occur is when the transmission line and/or tube is
terminated in its characteristic impedance.

Here is a simple explanation to the rescue, imagine the following, ten
nickel, spaced half an inch apart, lined up in a straight line, a drop
of glue is placed  on the center of each nickel, and a long rubber band
is laid down over the nickels, if you now accelerate the firs nickel
into the second, the energy (sometimes in the form of a forward moving
nickel and sometimes in the form of high pressure nickel) will move
from nickel to nickel forward down the line, however once it reaches
the front nickel there are no more nickels ahead to allow for nickel
pressure buildup, so the nickel wont stop until the energy is turned
into rubber band tension, and now you can see that the energy will move
backwards down the line (sometimes, as before, in the form of forward
moving nickels, and sometimes in the form of tense rubber band
sections).

It's not reflecting off thin air.  It's reflecting from the open end
of the tube.

In the tube, the air is constrained to a column - the sound can't move
sideways through the wall.  When it reaches the end, the tube is gone
and the sound can expand sideways.  So at the end the sound changes
from being plane waves to being spherical waves.

That process is not 100% efficient (changes always cost you) so not
all the energy goes into the spherical wave.  The part that does not
go into the spherical wave has only one place to go: back into the
tube.

Ken Plotkin