Nightclub physics: Fourier decomposition with cocktail glasses

[H2Bro]

This is purely conceptual and I'm just looking for opinions on whether its misguided or, indeed, plausible.

From what I understand about Fourier decomposition we can break down an analog signal into component sinusoidal waves. My thinking is that the sound system at a nightclub can be considered such an analog signal, and we can crudely/roughly get a decomposition of the activity of various wavelengths by watching excitations in the surface of water placed in glasses of varying diameter.

To demonstrate this I would just line up say, a shot class, a tumbler glass, a highball glass, and a martini glass each filled with water, or alternatively several glass shakers each filled with water to a different level, and point out how each one vibrates at different points to the music which corresponds to the different frequencies.

Would this be a correct interpretation of the phenomenon? Likely in actuality the heavy bass would just make them vibrate all at once, but hey, I'm a bartender, not a physicist.

 

[davidm1732]

There's a device called a Frahm tachometer that I believe kind of works on a similar principle...it has a series of reeds each with a different resonant frequency. When exposed to a complex vibration signal, the reed vibration will indicate what tones are present in the signal.
With your idea, all the glasses would be excited by the complex signal (not just the bass tones). You would need to know the resonant frequency of each device (glass), and note when the amplitude of vibration in the glass was at its peak...that would indicate that your source signal had a pure tone component equal or close to the resonant frequency of the glass.
Practically, resonance takes a bit of time to build up, so transient signals like music would be hard to map with such a system...and if you're a good bartender...the liquid wouldn't stay in the glass for very long!

 

[mikeph]

There is lots of complicated maths going on, but as long as you calibrate the glasses to find their frequency responses beforehand I don't really see any reason why you can't use this to actually make an estimate of the spectrum.

 

[K^2]

There are many problems with this. First of all, Fourier Transform gives you amplitude and a phase for each frequency. With glasses, you only get an amplitude. The other part is that excitations in a glass eventually die out. So "old" information about signal is lost. This means you have Fourier Transform over a window, rather than that of the whole signal.

There is a lot of information you can derive about the signal from resonances, and indeed, there are many hardware and software implementations that do just that. But you cannot treat it as just being a Fourier Transform of your signal. They are not the same.

 

[H2Bro]

Thanks for your responses! I think perhaps I misapplied the concept of a Fourier Transform (although I mentioned I'm going for a Fourier decomposition - are they the same or different?).

One thing to note is I'm not literally trying to make a Fourier transform, rather, I'm intended to show a bit of "bar physics" and just to get people thinking about what's going on with the sound they are hearing, how this can show up in a simple demonstration with glasses of varying size. Not a lot of thinking goes on at nightclubs and I'm trying to consider ways to improve that ;)

MikeyW, Could you elaborate on what you mean by calibrating the glasses to find their frequency responses? would this mean only having for example a thin layer of liquid at just the specified diameter as opposed to a tapered glass that is full of liquid up to a point?

Now I'm trying to wrack my brains to think of a method of recording the high amplitude responses through time so its not just a snapshot... If a wire was submerged in the glass would its capacitance change noticably as the water level moved up and down?

 

[K^2]

I would not use water level as indicator. That will introduce other harmonics. You need a way to measure movement of the glass itself.

 

[H2Bro]

Would a quartz or crystal glass emit peak electrical discharges that signify peak amplitude resonance, or would the sound be insufficient to activate the piezoelectric effect?

 

[K^2]

Piezoelectric effect requires crystal lattice. Glass, even glass commonly known as quartz or crystal glass, is still amorphous. There is no piezoelectric effect in any kind of glass.

You might be able to attach a small piece of quartz crystal, however, that will minimally affect the resonance frequency, but allow you to take the measurement. This would effectively act as a piezoelectric microphone. Some other kind of microphone could also work.

 

[mrspeedybob]

...
Not a lot of thinking goes on at nightclubs and I'm trying to consider ways to improve that ;)

This may be a bit off topic, but I just had to respond to this statement.

I believe a great deal of thinking goes on at night clubs, it's just that very little of it is about physics. Most of it is on psychology and behavioral science. Questions are pondered such as...
"How can I get that pretty girl to come home and have sex with me?"
or...
"What is the best way to wiggle my bottom to get that cute guys attention?"

Perhaps you would have more success in turning your customers' thoughts toward scientific matters if you tried topics on which they were already intently focused ;-)

 

[H2Bro]

Your right, a great deal of strategizing goes on in nightclubs. The funny thing is that all the strategizing is being done by intentionally self-intoxicated people. It's like preparing for a chess round by getting your head punched in, which incidentally is a sport called chess-boxing.