Physics Experiment - building a speaker

[JasonDelta]

Recently I've been toying around making my own speaker using a magnet wire coil and attaching it to the bottom of cups and bowls and then connecting it to a receiver for a simple home-made speaker. What I'm interested in doing is analyzing how well each material reproduces sound, so I was thinking about trying different materials and using a microphone to see how well different individual frequencies are reproduced. In order to compare the different materials to each other I was going to compare materials of differing elastic moduli. I was planning on setting it up fairly simplistically, by removing the bottom of a paper cup and then taking sheets of different materials and placing it over the end of the cup and then attaching the coil onto that. The only problem I'm wondering about is that there are other factors to deal with in each material, like thickness. Right now I'm not too sure where to go with this, but how does this sound so far? Do I have the beginning of a good research project, or is this just repetitive stuff that is already readily known?

As for the frequency reproduction, I was thinking about testing ranges of frequencies, from the low end (like 100 hz) to the high end (10,000 hz) with regular intervals in between. I'm planning on connecting a laptop to the receiver so that I can run a frequency generator program.

Thanks for your help in advance, I truly appreciate it!

[SY]

This is stuff that's been pretty well known for many, many years (like 50 or more). JAES and IEEE journals will turn up hundreds of papers. But I would encourage you to do it anyway (very educational!), and maybe try a few nonconventional materials that aren't in the literature.

Also, you might consider impulse or MLS testing, rather than continuous sine wave.

[sreten]

A good research project ? I'm sorry but I don't think so,
because you'll find great difficulty obtaining a coherent
analysis of your results, for one example the cube of the
thickness is critical.

There are a myriad of other factors to take into account,
neglecting any one of these will render the results essentially
meaningless.

Finding an area of original research in loudspeaker design is very
difficult and the physics involved can become very daunting, as
well as the practicalilities of obtaining accurate measurements.

Modern methods use finite element analysis for modelling and
laser inferometry for measuring moving element behaviour.

sreten.