Optical DIY mic idea

  • Thread starter Thread starter dres
  • Start date Start date
D

dres

New member
I've had an idea for a while for an optical mic design... but I suspect it's been done before ?

In my work I use this little photoamplifier. Only cost about $20 US and contains a lens, photodiode and a little pre amp. They have huge bandwidth (DC-MHz) and seem to have ample resolution. Solid state lazer diodes are also readily available these days..........so I was thinking....

Relfect the lazer light off of a thin reflective surface and onto the photoamplifier. Movement in the surface through sound waves should cause the laser light to move and give you a usable sound signal from the photodiode ?

I've seen similar ideas used for bass guitar pickups. Does anyone know if there are mics designed in this way ? Worth $40 to experiment ?
 
You are way beyond me, but I think this idea sounds incredible. If it excites you to do it, I'd say its worth $40 to try! The All Star Game tickets were $175 and nobody even won!
 
Yes! I've had the exact same idea for quite some time but never seem to have the time to pursue it... always seem to get involved in some other 'big idea' of the moment. Another application I considered for this was to use the technique as an acoustic guitar pickup by placing a small piece of reflective tape on the under side of the guitar top & reflecting a laser off it. One disadvantage in that application that has come to my mind is that the laser would be picking up just a single point source of vibration in the guitar top which is generally why any acoustic contact pickup for guitar sounds inferior to micing it. I seem to remember one of the major mic companies having a mic based on optics years ago but the memory is vague & could have been a dream :confused: . Maybe Harvey has seen such a thing.

-SD Simpson-
 
Unless I'm missing something the hardest part about the concept would be isolating the opto cell. It would have to be in a black box with a hole to allow sound to enter. But the hole would mess up the sound waves.

I guess you could always use a blacked out vocal booth and give the vocalist night vision goggles.

It is an interesting idea. I could see it having more value as a triggering device rather than an actual mic.
 
TexRoadkill said:
Unless I'm missing something the hardest part about the concept would be isolating the opto cell. It would have to be in a black box with a hole to allow sound to enter.
Click to expand...

No. That's not a problem at all. The whole thing could easily be enclosed & isolated from external light. The thin vibration membrane itself can be part of the external light shield (cover or whatever). Sound waves can strike it on the outside surface while the laser & photodiode reflect & detect off the inside surface. There are other technical design problems to surmount but that's not one of them.

I have some question as to what significant advantage there would be. One thing I see is it would allow for an extremely low mass small diaphragm & compliant suspension, which would allow for extremely wide bandwidth (ultrasonic even) if you could devise a proper mounting & suspension of it that wouldn't be to fragile.

-SD Simpson-
 
There were some guys at Vdrums.com who had a similar idea a while back for use as a drum trigger. I don't know how far they got, but you might want to check out the forums over there and bounce ideas off these guys. Sounds like a fantastic idea to me.
 
How much resolution would you get from the laser light? I would assume that the light would have to be equal size to the membrane to accurately reproduce the sound? Or would you just have to hit the center of the membrane? I guess a laser and a photoamplifier seem, to me, a little to discrete to reproduce a sound wave.

Then again, you would basically be measuring the amplitude of the membrane in terms of voltage. So, yes, I've talked myself back into it. You just have to hit the center.

I agree that the issue would not be neccesarilly in the laser/photoamplifier, but a membrane that accurately reflects the light to the photo amplifier.

You could also set the laser to pulse at 96 kHZ (or whatever you want) and use a data aquisition routine on the photoamplifier that is synced at 96 khz and create the worlds first digital mic! Stick a tube in that you analog weenies!
 
JohnWaynesTeeth said:

You could also set the laser to pulse at 96 kHZ (or whatever you want) and use a data aquisition routine on the photoamplifier that is synced at 96 khz and create the worlds first digital mic! Stick a tube in that you analog weenies!
Click to expand...

Now *that* is a brilliant idea !!

Actually I did a little surfing on the web. Turns out it's not a very new idea as I suspected. Sennheiser have done R&D in this area as well as others. search for "optical microphone" and see what you get.

Apparently there are surveylance methods that use this exact principle. Basicly you can bounce the lazer light off a window from the outside and eavesdrop on conversations inside !! You would think the mass of the window would kill your frequency response but apparently it works quite well.

For my experiments at work I did try to initially build an preamp circuit for the el-cheapo photodiodes you can get. Unfortunately the noise was pritty enourmous. Then I spied these photoamplifiers in an RS catalog... they have a preamp circuit built into the casing with the photodiode and are virtually noise free. They can run off a single rail supply (+5 up to +15V) and the output is in the same range (ideal line level). I use one with a bandwidth of 65kHz but if you pay 4x the price you can get high speed ones that go up into the GHz range !

As far as resolution goes I use mine at work with a 16bit A/D card and theres no sign of any significant noise even down to the level of the cards quantisation.

You'd need a fairly accurate mounting arrangement. I suspect ambient light might not even be a big problem since if it was constant would only introduce a DC offset. It might be possible to build something the size of a matchbox you could place near any reflective surface to turn it into a mic ! The possibilities are endless ! Could be a money spinner for someone with buisness skills ? (ie not me ;) )
 
Last edited:
Semheiser is exactly the major mic company I thought I remembered seeing with something like this years ago but wasn't absolutely sure it was them. I'm quite aware also of it's long time use for snooping as you described by bouncing the laser off a window. In fact this thread had me kind of looking through my stacks of old Popular Science/Mechanics/Electronics/etc.. magazines for the one I know I had at least 10 years ago with an article & DIY project for building one of those window bouncing laser snoopers. This thread also reminded me I really need to throw away a lot of magazines :D . I agree there still appear to be many untapped possibilities with the concept though.

Cheers :)

-SD Simpson-
 
knowdoubt said:
I have some question as to what significant advantage there would be.
Click to expand...
The real advantage would be if the optical mic overcame the dynamic range limitations of traditional mics.
 
I also think there are real advantages in being able to seriously tinker with the acoustic qualities. It opens up all types of possibilities for colouring the sound by using different materials for the reflective membrane... how about a plastic coke bottle mic ? Maybe it would sound awful ? Maybe it would be just the kind of inspirational "spice" you need to use as an effect ? Regardless, it would be kina fun to try.
 
dres said:
I also think there are real advantages in being able to seriously tinker with the acoustic qualities. It opens up all types of possibilities for colouring the sound by using different materials for the reflective membrane... how about a plastic coke bottle mic ? Maybe it would sound awful ? Maybe it would be just the kind of inspirational "spice" you need to use as an effect ? Regardless, it would be kina fun to try.
Click to expand...

Absolutely. One could say the central, most distinguishing feature of this technique is that the transducer mechanism that converts the acoustic energy to electronic output is so physically isolated & independent of the acoustic element, placing absolutely no physical load on the acoustic element (or object) & very little constraint on it's properties or it's physical proximity to the electronic conversion apperatis. About the only constraint is that the acoustic membrane or object must be somewhat photo reflective or have a photo reflective attachment. The acoustic element is then free to be nearly anything, even the source sound producing object itself (such as a guitar top). 'Plastic coke bottle'.... :) I love that, good illustration of this freedom.

TexRoadkill ;
Yeh... I'm not sure what the dynamic range constraint of the optical portion would be, but that is a worthy consideration. The acoustic element of course is free to be just about anything from a dust particle to a wall so the optical portion is the question I guess.

-SD Simpson-
 
Will you be trying to amplitude modulate the light with the membrane, or do something like interference with the membrane the moving mirror? Just curious...
 
Interferometry would be the common approach since vibration of the membrane would not directly influence the amplitude of the reflected beam.
 
knowdoubt: you'd need a beam splitter to do interferometry wouldn't you ? what could you use for this - ordinary glass ?

I was actually thinking of employing a grazing angle method - as this is what sennheiser have tried.
 
Your right. The grazing method is by far more simple & less costly. That's actually more the way I initially envisioned doing it a few years back (Duh.. it's been awhile since I've thought of any of this). A grazing method does modulate the amplitude of light detected since the movement of the membrane causes a greater or lesser % of the beam to strike the detector. Forget about interferometry.

Sennheiser Optical Microphone
Seems like it would be a piece of cake DIY project to build a decent working mic using Senheiser's method shown in the link above doesn't it?! The fiber optic's is the beauty of it, focusing the light into a beam of controlled cross sectional dimension. With that I should think a simple LED could be used.

Well, wait a minute, not so fast... thinking off the top of my head now. Is there going to be any question of linearity here? I mean.. the amount of change in light energy has to be such that the end result is a voltage change directly proportional (with no distortion) to the displacement of the diaphragm. Now, with the grazing method shown above I would assume it is adjusted so that when the diaphragm is at rest (total silence) the % of the reflected beam striking the detector is something greater than or less than 50%. Bias point for maximum dynamic range I guess would be either 75% or 25% but I don't guess the diaphragm is going to move enough to worry about biasing it at the center of the dynamic range, as long as the beam doesn't cross the 50% barrier ( move from above 50% to below it or visa versa) or move totally off the detector or cross the point of 100% detection. Anyway, 'is a circle the right shape to give a linear transfer function'?? I guess so... I don't know for sure. Any idea?
 
Linearity is the big issue I suspect... you might be able to check it using a micrometer to physically deflect the membrane by known distances. But for DIY people like us I think acoustic colour might be the magic of such a device.. in which case non-linearity might be a good thing ! (ie tubes)

So are we going to just talk about this or make one ? ;)
 
I'd love to make one. Unfortunately that would require me to get up off my ass. :D . Seriously though, you've re-sparked my interest & if I can find any time I do intend to give it a shot. I just don't have much free time for such things at the moment & I seem to have less & less free time as time goes on. I'm just getting started on a pretty big recording project that will take most of my free time attention for awhile but I may be able to do a little here & there on the mic. I hope you follow through with it & keep me updated with any progress & results & I will do the same. We may each have a suggestion or 2 along the way to help the other. I'd seriously like to keep in touch on this. Email would be best as I may miss a forum post.

Do it! :)
 
knowdoubt: can you send me a private message with your email address ? Thanks

I'm going to contact IPL and see what might be the best place to purchase these little photoamplifiers.
 
You must log in or register to reply here.
Back
Top