There is a book called _The Theory of Sound_ by some guy named
Rayleigh.
If you haven't read this lately, you've got to get a copy and
read at least the first 100 pages. Every two or three pages
will give you an idea for a simple but enlightening experiment
to do. The title of the book says _Theory_ but Rayleigh was
also a past master experimentalist, with a taste for simple,
elegant, incisive experiments.
Example: Lissajous patterns. Rayleigh (and of course Lissajous
himself) did all sorts of clever things with Lissajous patterns.
Coincidentally, when I was in high school, I spent hours making
Lissajous patterns on the oscilloscope. I say "coincidentally'
because at the time I hadn't read Rayleigh and had not the
slightest inkling that what I was doing was useful. It was
play, not science, just a hobby, just a diversion, i.e. just
something I could do during physics class and not get into
trouble for.
Over the years since then, in the research lab, I've used
Lissajous patterns again and again. Recent example: plot
the current versus voltage for a piezo transducer hooked to
a complicated mechanical thing, in order to help find and
understand the horrible electro-mechanical resonances.
For some kids, it might be OK to just send 'em into the lab
with instructions to put up an "interesting" Lissajous pattern
... but probably you want something not quite _so_ open-ended.
So here's a partially-baked more-specific suggestion: give
them two tuning forks and tell them to add bits of tape or
wax or solder or whatever so as to make them exactly one
octave apart, as exactly as they possibly can. Hint that
microphones are available and that the scope has an XY mode.
Or make a siren -- i.e. a rotating disk with holes -- to
demonstrate the relationship between sound and periodic
disturbance of the air. Add in a small fraction of a
_subharmonic_ to illustrate Rayleigh's point that "pitch"
cannot be defined precisely. Do the perceptual experiment
to see how much of the subharmonic can be added in before
the note begins to be perceived as being an octave lower.
Or the Lissajous bars: Two pieces of bar stock joined
end-to-end _crosswise_ so that the X motion is virtually
independent of the Y motion.
Or a Blackburn pendulum.
And so forth. You get the idea. Read Rayleigh already.
Here's another open-ended sound/wave related idea: the task
is to use basic physics (not an oscilloscope) to demonstrate
that sound is a wave. Hint that the idea is to draw the
waveform on the wall. Cat lasers (i.e. dollar-store lasers)
and turntables are provided. Quarter-mil aluminized Mylar
is provided. For once they are encouraged to bring cell
phones and/or iPods to class. Even with that level of hint,
the project is still quite open-ended; there are many
variations and many design decisions that must be made.
Also: Ripple tanks! In high school I spent many many
hours experimenting with ripple tanks. This was not in
the same category as playing with Lissajous patterns; I
knew this was serious physics. I knew enough about
diffraction and refraction and interference to be able
to set myself some quite specific goals (not all of which
were met). There's a lot you can do with a ripple tank.
We had 30 ripple tank kits, which was at least 25 more
than were needed, so I cannibalized three of them to make
an über-fancy version. Beware: there's also a lot you
/cannot/ do with ripple tanks. For one thing, the water
wave equation is highly dispersive, which makes it a
lousy model for sound or for light unless everything is
monochromatic. On the third hand, knowing about dispersion
is plenty valuable. Waveguides are dispersive. Not
coincidentally, the wavefunction for a massive particle
is dispersive. Even sound and light are dispersive in
polar coordinates. (Bessel functions are not the same
as plane waves.)
In case you're wondering, my entire high-school physics
experience was open-ended to a fault. By the time I
got around to taking HS physics, I had already taken
college math in summer school and night school ...
calculus and beyond. Also I had read Feynman volume
I cover to cover. The teacher had never heard of
Feynman AFAIK. So you can imagine that I didn't spend
a whole lot of time paying attention to what was going
on in class. The teacher took pity on me and let me sit
in the back and read ... or play with the scopes, ripple
tanks, vacuum pumps, et cetera.
One more thing: An open-ended assignment that may seem
trivial, but isn't: Assign them to make a giraffe out
of baling wire. Just a simple one, a few inches high.
The point is that despite all the clichés about baling
wire, kids don't have a clue about how to use it. The
artistic standards for the giraffe are minimal: It needs
to have four legs, a short tail at one end, and a long
neck at the other end. The crux is that it is required
to be sturdy. (If you make it in the obvious amateurish
way, it won't be.) Let 'em fail, then show 'em how to
do it right.
This is relevant because in the real world, when you're
doing open-ended research, there is a lot of improvisation.
Maybe 90 or 99% of your hypotheses will fail, so it is
important to have a high "attempt frequency" (as Arrhenius
would put it). Rapid improvisation. Rapid prototyping.
This reminds me of Mythbusters: When those guys pretend
to be scientists, I don't know whether to laugh or cry
... but one thing you can say for them, they've got
great improvisational fabrication skill -- which is a
valuable thing.