Re: mic for low frequencies

[Trebor]

What about an optical solution: bounce a laser(pointer) off the surface of the oscillating liquid and monitor the intensity of the reflected beam with a photocell or phototransistor. laser microphone style, (but not interferometric).

Or possibly stick a magnet to your apparatus and position the head from a tape deck close to, but not touching it, You'll have to make your own pre-amp though as audio preamps will have filters to remove sub-bass ...

Rumble filters are high-pass filters applied to the removal of unwanted sounds below or near to the lower end of the audible range. For example, noises (e.g., footsteps, or motor noises from record players and tape decks) may be removed because they are undesired or may overload the RIAA equalization circuit of the preamp.

http://en.wikipedia.org/wiki/High-pass_ ... plications

[kozikowski]

<<<Or possibly stick a magnet to your apparatus and position the head from a tape deck close to, but not touching it, >>>

That won't do DC. Moving coil or moving magnets are all velocity systems. The faster the movement, the larger the output signal, and they have to be specially compensated for that if the application is audio.

Ever meet a Theremin?

http://en.wikipedia.org/wiki/Theremin

Those will create a sound signal against the position of your hand or other conductive object. The volume of the instrument is one paddle and the tone pitch the other. The system will transmit DC. If the output tone rises from 1000Hz to 2000Hz and stays there for example, that means an object has approached the pitch paddle and stopped moving. If the object never moves, the output tone will remain at 2000Hz or until something breaks or goes out of alignment. 2000Hz tone represents a DC level and designing pitch to DC converters isn't that complicated.

The usual Nyquist restrictions apply. The lowest pitch tone needs to be higher than 2.6 times the speed of the object. So 10 vibrations or position changes per second needs to be carried by output pitch tones no lower than 26Hz for any kind of accuracy. I'd go higher.

This design has the advantage that you can record the changing pitches as regular, standard audio signals in a WAV file and convert them into motion and position information whenever you want.

Koz

[kozikowski]

OK, lets take this mechanically one at a time. How do you know it's oscillating? What changes? Is there a meter somewhere that goes up and down. Pressure gauge? Color change? Can you feel the table moving?

Koz

[kozikowski]

I have a minor history in radio transmitters and they're famous for flying blind. You can't actually measure anything, so you have to measure oblique effects and derive the information. Like you know you succeeded in creating radio power when the electrical meter outside the house spins a little faster.

That's a silly example, but it's not that silly. Particularly if the system has to respond to zero, you may have some really entertaining calibration processes. Detecting motion or changes is relatively easy. Detecting when a system approaches zero is a lot harder.

Koz

[Trebor]

<<<Or possibly stick a magnet to your apparatus and position the head from a tape deck close to, but not touching it, >>>

That won't do DC. Moving coil or moving magnets are all velocity systems. The faster the movement, the larger the output signal, and they have to be specially compensated for that if the application is audio.

What I described is similar to a geophone, which allegedly go down below 10Hz ... http://en.wikipedia.org/wiki/Geophone

But I take your point that the signal intensity would fall off rapidy at lower frequencies.

Your theramin idea is interesting, and not frequency dependent, but to expand upon my original optical suggestion, sticking a mirror on the apparatus and using intereferometry would be a more reliable non-frequency-dependent method than a theramin: less likely to pick up taxi cab radio messages ...

Environment can play havoc with any theremin; EMI electro-magnetic-interference from fluorescent lights, microwaves ovens, computers, AM radio stations, etc, are notorious for ruining a playing session.

http://www.oldtemecula.com/theremin/index2.htm

[kozikowski]

Yes, they did work on multiple RF oscillators, but the concept is to change the event into something you can easily measure and then record that instead of trying to turn the event into a varying battery signal right away.

I don't think anybody fully appreciates the significance of the specification. Say you leave your electric torch (flashlight) on and unattended by accident. A couple of hours later you will need a trip to Tesco/7-Eleven to restore the flashlight (electric torch) to operation.

That wasn't just an unfortunate accident. That was the show.

Send me a copy of it please. I'll give you an FTP site to use.

Thanx,

Koz

[Storer]

Sounds to me like a variable rate stroboscope might do the trick, as long as there is some way to make the oscillations visible.

[kozikowski]

<<<Sounds to me like a variable rate stroboscope might do the trick, as long as there is some way to make the oscillations visible.>>>

Talk about that. Shine a strobe light on the liquid? I think it's valuable to solve the DC "unmoving" problem first since that's the hard one. So there are no oscillations. The liquid has stopped in one state.

Koz

[Irish]

Are we talking about a fluidic oscillator like the one in the top pictures here? http://aarls.eng.ohio-state.edu/PSP/fluidic.html

Then how about this for a solution? Put a point light source on one side of the jet and two photocells (or, more likely, phototransistors) on the other side, so that the light shines through the jet onto the photocells. As the jet oscillates from side to side, the light is refracted through the fluid, and falls more on one photocell than the other if the jet is further to one side than the other. That way, you get both steady-state and dynamic output by comparing the output from the two photocells.


PO'L

[kozikowski]

That's more better. Although you would think the curve was a straight line with quiescent at the center, it's actually an "S" curve. Suppose the sensing beam overloads the system and overshoots the cell?

http://www.radio-electronics.com/info/r ... lation.php

"S-Curve" About two or three pages down.

OK, you still can't get that into Audacity. How about two DC to Frequency converters, one on one photothing and one on the other. Put one on Audacity left and the other on the right. That lets you record the event as standard audio tones.

Or how's bout send two tones into the photothings. The photothings will change the sizes of the two signals versus light.

It's a given that all this must be strictly temperature controlled. All these devices drift with temperature.

Koz