Monochrome interference microphone

[poor mystic]

Idea for a microphone.

Hi People!
I've had an idea for a monochrome light microphone, which I have rendered as a drawing, attached.
The detector consists of a monochrome source (l.e.d.), a reflective membrane (gold leaf) some distance from the source, a static reflective element, and several randomly-placed detectors, 3 of which are shown.
The idea is that as the membrane moves towards or away from the plane of the source and detectors, each detector will be excited at different times, when constructive interference exists between the light reflected from the moving membrane and the light reflected from the static reflectors.
As a sound wave pushes the membrane towards the detectors, the detectors will tend to be excited in some characteristic repeating sequence, I hope; the order will be reversed when the membrane is moving away from the detectors.
With 'towards' and 'away' discrimination provided by the sequence discriminator, a count can be incremented or decremented respectively. The count therefore shows the position of the gold leaf.
Questions & Problems
First, I wonder whether this has been tried before, and then I'm looking for any ideas on how to train some kind of sequence discriminator to recognise "toward" and "away" movements of the membrane.
The biggest problem is the light. I think I'd be a lot better off using a low infrared monochrome source than a visible light source. This is because there are so horrendously many wavelengths of visible light per millimeter. It would be better to use a wavelength of hundreds of nanometers, if practicable.

Still, it's just an idea at this stage... any thoughts?

 

[berkeman]

Idea for a microphone.

Hi People!
I've had an idea for a monochrome light microphone, which I have rendered as a drawing, attached.
The detector consists of a monochrome source (l.e.d.), a reflective membrane (gold leaf) some distance from the source, a static reflective element, and several randomly-placed detectors, 3 of which are shown.
The idea is that as the membrane moves towards or away from the plane of the source and detectors, each detector will be excited at different times, when constructive interference exists between the light reflected from the moving membrane and the light reflected from the static reflectors.
As a sound wave pushes the membrane towards the detectors, the detectors will tend to be excited in some characteristic repeating sequence, I hope; the order will be reversed when the membrane is moving away from the detectors.
With 'towards' and 'away' discrimination provided by the sequence discriminator, a count can be incremented or decremented respectively. The count therefore shows the position of the gold leaf.
Questions & Problems
First, I wonder whether this has been tried before, and then I'm looking for any ideas on how to train some kind of sequence discriminator to recognise "toward" and "away" movements of the membrane.
The biggest problem is the light. I think I'd be a lot better off using a low infrared monochrome source than a visible light source. This is because there are so horrendously many wavelengths of visible light per millimeter. It would be better to use a wavelength of hundreds of nanometers, if practicable.

Still, it's just an idea at this stage... any thoughts?

One variation of your idea that exists today for position measurement is the laser interferometer systems like these from Agilent:

http://www.home.agilent.com/agilent/product.jspx?nid=-536900386.0.00&lc=eng&cc=US

So instead of using multiple detectors to figure out movement, you use the changes in intensity from constructive/destructive interference to tell you how many wavelengths the distance has changed.

You might be able to make a small version of this for your microphone, using a laser diode. I'm not sure what the coherence length of a typical low-cost laser diode is, though. You could use a laser pointer's diode for initial experimentation (but please wear laser eye protection if you start working with laser diodes).

 

[poor mystic]

Thanks Berkeman!
Still, I'm sure the professional systems use more than 1 detector, otherwise they wouldn't be able to detect the sense of any movement - whether it is towards or away from the detector.
I've made a small advance in understanding. If just 2 detectors are 1/4 wavelength apart the order in which they detect the changeing sums of interfering waves shows the sense of the movement.

 

[poor mystic]

Further question:
I've been turning this over and over in my mind and I just don't see why a laser should be required. It seems to me that a Young's double slit should work just as well, with no dangerous lasers. Perhaps I got that wrong?