Print PDF
Favorite
Email
There are many articles on the web about making condenser microphones. These projects are usually relatively easy. However, trying to make one that allows X-Y and Wide Stereo positions for the mics that some professional recorders offer (and in an inexpensive way) may seem to be a difficult task.
This project has been in my mind for a couple of months and I have finally reached a simple yet (cost-) effective solution. I would like to share my steps of making it and hope that someone out there would find this instructable useful.
This project has been in my mind for a couple of months and I have finally reached a simple yet (cost-) effective solution. I would like to share my steps of making it and hope that someone out there would find this instructable useful.
Remove these ads by
Signing UpStep 1Mounting the Condenser Mics
Parts:
6.3mm Right Angled Plug x 2 (I took them off from a fender patch cable)
Panasonic WM-61A Electret Condenser Microphone x 2
Silicon Earphone Bud x 2
Some Silicone Sealant
Instructions:
It is easy if you have the right parts of the right size.
- Put the silicon earphone bud to the end of the right angled plug. It should be a tight fit as the bud is slightly smaller than the end opening of the plug.
- Apply some silicone sealant to the inside of the plug to secure the bud.
- Solder the wires to the mic and to the connectors of the plug. Remember to run the wires through the bud and the plug housing before soldering (which I often forget to do...).
- Insert the mic to the bud and it should be another tight fit. Remember to push the wires with some tool in instead of forcing it in by pushing the mic, or you may break the soldered joints.
6.3mm Right Angled Plug x 2 (I took them off from a fender patch cable)
Panasonic WM-61A Electret Condenser Microphone x 2
Silicon Earphone Bud x 2
Some Silicone Sealant
Instructions:
It is easy if you have the right parts of the right size.
- Put the silicon earphone bud to the end of the right angled plug. It should be a tight fit as the bud is slightly smaller than the end opening of the plug.
- Apply some silicone sealant to the inside of the plug to secure the bud.
- Solder the wires to the mic and to the connectors of the plug. Remember to run the wires through the bud and the plug housing before soldering (which I often forget to do...).
- Insert the mic to the bud and it should be another tight fit. Remember to push the wires with some tool in instead of forcing it in by pushing the mic, or you may break the soldered joints.
| « Previous Step | Download PDFView All Steps | Next Step » |
I love the idea of using the right angle jacks to give you pivoting capsule holders, though. You might look into using transound cardioid capsules instead (see http://www.transsound.net/) - the TS-55A looks like the smallest with reasonable frequency response. It's 9.7mm in diameter though, and you must provide an acoustic path to the back of the capsule for the directionality to work, so you'd need something more elaborate to mount the capsule at the end of the jack.
The coolest setup would be if each pivoting arm was long enough and positioned such that when pointed towards each other at 45 degrees they just touched (each pointing across the face of the other) giving you a classic X-Y coincident, while if pointed away from each other at 110 degrees they ended up 17 cm apart giving you a classic ORTF near-coincident setup. Super-cool would be if each arm could telescope some too, which would really let you experiment with any variation.
http://www.soundonsound.com/sos/apr98/articles/mic_types.html
http://en.wikipedia.org/wiki/Microphone#Microphone_polar_patterns
The WM-61 capsules are omni-directional as you mentioned. I wasn't concerned about this when I built this thing because many others are using these capsules for the same/ similar purpose and have reached satisfactory results. This is one of the pages I read before building this:
http://www.subatomicglue.com/mintymic/
However, your post really got me into thinking about how and why/ why not the polar patterns matter. I think the fact that others have achieved good results with these capsules is first because of these mics' frequency responses and sensitivity, then the acoustics (how the capsules are installed, in what direction/ angle, materials used in the holder, etc) in their builds.
I think our ears are more omni-directional than directional, and they are separated in between by our head, so your suggestion to have a sound baffle in between the mics makes a lot of sense to me and can be easily implemented with the current setup.
The TS-55A has a much lower impedance and maybe I cannot just replace the mics and try them out. If I am more into this, I should add some switches on the box with different resistance value built-in so it is possible to match the impedance and try out different mics/ builds of the mics.
My current project is to build another set of mics, with adjustable (in both length and angle) pivoting arms. This is for a friend who shoots videos and should allow him to set the mics a bit closer to the objects without having a boom operator. Your last paragraph has given me some good ideas in the design of this build, which I think can be incorporated into my current project and make it serve more useful under different circumstances.
Testing and practical use will depend on how you actually want to use the microphones. There are numerous schools of thought with regards to creating a good stereo image, and there are many factors to consider that are very closely related to one another, particularly the phase relationship between the mics - this determines mono compatibility, proper imaging without a "hole" in the center, comb filtering that colors the sound, and so forth. The distance between the two mics, their relative angles, and polar response are major determining factors here.
I would experiment by figuring out the most common relative angles you might use (meaning each mic is aimed at half the total angle from the center): 90 degrees, 120 degrees, 155 degrees (which is the correct angle for rear speakers in a 5.1 setup), and 180 degrees. With each, I would test both toe-in and toe-out configurations of each. Measure the distance between the center of each element, then create a testing area in a large room with plenty of space to work with and good absorption (like a living room - couches, curtains, and carpet are free reverberation control). Place the mics in the center of the room, and mark out arcs at .5 meters, 1 meter, and 2 meters.
Next, I would get a portable sound source with the best audio quality I could find (a portable radio with a single speaker might do in a pinch) and, using each arc as a guide, play the audio at 0 degrees, 45 degrees, and 90 degrees (you can assume symmetry for the other side, so there's no redundancy). I would also do a final test at each distance by gradually moving the sound source along the arc to see if the imaging sounds smooth and without holes.
If you really want to get precise, let me know - I'll hook you up. For that matter, I'll even analyze the results for you if you like, as I'm curious to know myself.
I can understand the idea of recording with a portable sound source at different angle and distance, and moving along the arc. Then I can compare the results with the mics in different settings (toe-in, toe-out, both pointing straight, pointing to opposite sides, etc.). I wonder what kind of software I need to analyze the results though. This is actually something I may try in future and I would like to learn more about it.
My original plan was to take this thing out to some gigs and compare different settings just by ear. I have the sense that the size of the venue should also determine the "best" setting that should be employed.
With this little thing, I can also build another set of mics for Binaural Stereo Recording. Something like these should work (glenglenn.com/index.php/binaural-stereo-recording/). It should be easy 'coz the jacks are already built-in there.
If you imagine the unit is your head, a straight line drawn from back to front will, of course, be 0 degrees. For the first test I suggested, you would toe both mics either in or out by 45 degrees, making their total angle (relative to each other) 90 degrees.
I recommend measuring the distance between each mic because this will help you calculate any comb-filtering that will color the sound, cause mono incompatibility, or other phase-related issues. Any audio frequency where the distance between the mics equals 1/2 the wavelength of the sound will cancel out. For example, if the speed of sound (at sea level) is 340 m/s, and the distance between the mics is 17 cm (they appear much closer in the picture, but this is an example after all) then the wavelength of the lowest frequency canceled is 17*2=34 cm, or 0.34 m. Since frequency is equal to the speed of sound divided by wavelength, 340/0.34=1000 - meaning that 1 kHz and all its harmonics will be canceled.
For subjective measurements, any sound source will do - although this means you cannot adequately analyze the results mathematically, it will at least give you a feel for the stereo image. A rigorous test would follow many of the guidelines I gave you, except it would be performed in an anechoic chamber with a speaker possessing close to flat response. As I'm sure you have neither at your disposal (as most don't), a large living room can suffice due to the fact that there are many absorptive surfaces such as carpet, drapes, upholstery, &c. For testing stereo imaging, you rarely will need to worry about frequencies less than 1 kHz (as anything lower wraps around the human head, meaning phase problems and directional cues really aren't perceived in that range) so a decent speaker like a surround will suffice.
As for the test signal, chirps (swept sines with rising frequencies between 20-20 kHz or better) are great; they're pure, broadband, mathematically easy to compute, and have excellent noise rejection during analysis. Most DAW software (like Nuendo, ProTools, and even Audacity) has some useful analysis tools. Even collapsing the .wav file to mono and performing a spectral analysis will yield some useful information. Using freeware tools, you can even create impulse responses that you can not only use for analysis, but also for reverb and other effects.
I apologize for being long-winded, but since you mentioned binaural recording I have to add one last bit. Binaural recording would use a similar setup to yours, but to pull it off properly you'll need to emulate the physics around the human head. Unfortunately, the only real way to do that is to make a dummy head. The head should be about the same size and shape as an average human head, and ideally should use dense materials like rubber or wood (although styrofoam can be used, it tends to resonate). The mics will be aimed at a relative angle of 155 degrees, and the elements should be just inside where the ear canal would be. Furthermore, you'll need rubber ears to complete the illusion as they are extremely important in helping us human listeners with directional cues (even the little ridges and bumps reflect sound in ways required by our brain to process direction). Build one, and you'll get great results that really "put you there" with a good set of headphones.
If this is too much, you could also craft a Jecklin disk to mount between the mics in your setup; while it won't necessarily mimic binaural recording, it will vastly improve stereo imaging both through headphones and loudspeakers.
With regard testing, I guess you want to record some things with different levels & positions etc?
L
L
L