I would like to build a shotgun microphone but cannot seem to find any information anywhere. (Not a parabolic type, for which there is much DIY info, but the conventional long tubular type used in tv/film.) Obviously, lots of design engineers working for established companies (Sennheiser, Azden, Audio-Technica, Shure, etc.) have it down to a fine art, but they're not talking.
Obviously, one could reverse-engineer a good microphone and fake it, but it would leave a lot of unanswered questions about the art/science behind these designs. For example, would one use an omni or cardioid capsule? What should the dimensions of the tube be, and what are the the trade-offs? How do those vents on the sides work? What controls the pickup pattern? The frequency response? Etc. Thanks much.
Scott Morrison
Van Nuys, CA
Obviously, one could reverse-engineer a good microphone and fake it, but it would leave a lot of unanswered questions about the art/science behind these designs. For example, would one use an omni or cardioid capsule? What should the dimensions of the tube be, and what are the the trade-offs? How do those vents on the sides work? What controls the pickup pattern? The frequency response? Etc. Thanks much.
Scott Morrison
Van Nuys, CA
Good question! The long tube has slots in it to cancel the rear waves. Figuring that out might be hard.
Years ago my uncle built a very strange "shotgun" mic.
It was about 3" around, and composed of a series of small tubes. Each was longer than the next, so you had a spiral barrel with the longest tube in the center.
It worked.....
Years ago my uncle built a very strange "shotgun" mic.
It was about 3" around, and composed of a series of small tubes. Each was longer than the next, so you had a spiral barrel with the longest tube in the center.
It worked.....
Thanks folks for the responses. Yes, I recall seeing the spiral tube array many years ago.
I agree that figuring it out (from scratch) might be hard, especially for a math-challenged guy like me, but I would assume there are some basic straight-forward design equations published that could be used to determine the physical characteristics of the tube (length, diameter, placement/size of slots, etc), which in turn would be dictated by variables such as desired frequency response, capture ratio, and so forth.
I would imagine equations would also exist for the spiral tube type (how many tubes, what length, what diameter) and the parabolic reflector type (size dish, focal point).
Meanwhile, the search goes on...
I agree that figuring it out (from scratch) might be hard, especially for a math-challenged guy like me, but I would assume there are some basic straight-forward design equations published that could be used to determine the physical characteristics of the tube (length, diameter, placement/size of slots, etc), which in turn would be dictated by variables such as desired frequency response, capture ratio, and so forth.
I would imagine equations would also exist for the spiral tube type (how many tubes, what length, what diameter) and the parabolic reflector type (size dish, focal point).
Meanwhile, the search goes on...
Hi Mark,
I really appreciate the Elektor directional microphone circuit offer. Yeah, I'd be curious what that's about. Do you have a general idea of what it consists of? Is the microphone portion of the unit a tubular shotgun type or a parabolic reflector type?
I should mention that the shotgun microphone information I am seeking is more mechanical in nature than electrical (i.e., vented hollow tube), other than what exists in the mic capsule itself.
I have a number of omni and uni-directional electret condensor capsules on hand, most of which I bought from Digi-Key for about $2 each. Ideally, I'd like to build my own poor-man's version of the tv/film "industry standard" Sennheiser MKH416 (>$1000) or equivalent if possible.
Thanks,
Scott
I really appreciate the Elektor directional microphone circuit offer. Yeah, I'd be curious what that's about. Do you have a general idea of what it consists of? Is the microphone portion of the unit a tubular shotgun type or a parabolic reflector type?
I should mention that the shotgun microphone information I am seeking is more mechanical in nature than electrical (i.e., vented hollow tube), other than what exists in the mic capsule itself.
I have a number of omni and uni-directional electret condensor capsules on hand, most of which I bought from Digi-Key for about $2 each. Ideally, I'd like to build my own poor-man's version of the tv/film "industry standard" Sennheiser MKH416 (>$1000) or equivalent if possible.
Thanks,
Scott
cyclotronguy said:
This one??:
http://www7.taosnet.com/f10/ShotGunMike.pdf
Frank /~ http://newmex.com/f10
@/
I wonder if that multi tube shotgun would do a better job with a different increment of length? Instead of doing every inch from 1 to 36, you could spread them out so each has approx the same bandwidth.
Make your longest one 1meter and reduce each successive one by 1.11940577580564 and you have 36 pipes that each share the same bandwidth (logarithmically spaced center frequencies) and vary in resonance from 172 to 10000 Hz.
There are more shorter pipes and it might make a more interesting "sculptural" effect.
Length, in Fres Length, mm
0.679 10000 17.2
0.760 8933.31 19.3
0.850 7980.4 21.6
0.952 7129.14 24.2
1.066 6368.69 27.1
1.193 5689.35 30.3
1.335 5082.47 33.9
1.495 4540.33 38.0
1.673 4056.02 42.5
1.873 3623.37 47.6
2.096 3236.86 53.3
2.347 2891.59 59.6
2.627 2583.15 66.7
2.941 2307.61 74.7
3.292 2061.46 83.6
3.685 1841.56 93.6
4.125 1645.13 104.8
4.617 1469.64 117.3
5.169 1312.88 131.3
5.786 1172.83 147.0
6.477 1047.73 164.5
7.250 935.97 184.2
8.116 836.13 206.1
9.085 746.94 230.8
10.170 667.27 258.3
11.384 596.09 289.2
12.744 532.51 323.7
14.265 475.7 362.3
15.969 424.96 405.6
17.875 379.63 454.0
20.010 339.14 508.2
22.399 302.96 568.9
25.074 270.64 636.9
28.067 241.77 712.9
31.419 215.98 798.0
35.170 192.95 893.3
39.370 172.36 1000.0
363.403inches 9230.4millimeters in total length of tubing
One side benefit is it uses almost half as much tubing....
I haven't built either one, but it makes intuitive sense to me that the resonant tubes should each share the same bandwidth, rather than be round numbers in length.
Make your longest one 1meter and reduce each successive one by 1.11940577580564 and you have 36 pipes that each share the same bandwidth (logarithmically spaced center frequencies) and vary in resonance from 172 to 10000 Hz.
There are more shorter pipes and it might make a more interesting "sculptural" effect.
Length, in Fres Length, mm
0.679 10000 17.2
0.760 8933.31 19.3
0.850 7980.4 21.6
0.952 7129.14 24.2
1.066 6368.69 27.1
1.193 5689.35 30.3
1.335 5082.47 33.9
1.495 4540.33 38.0
1.673 4056.02 42.5
1.873 3623.37 47.6
2.096 3236.86 53.3
2.347 2891.59 59.6
2.627 2583.15 66.7
2.941 2307.61 74.7
3.292 2061.46 83.6
3.685 1841.56 93.6
4.125 1645.13 104.8
4.617 1469.64 117.3
5.169 1312.88 131.3
5.786 1172.83 147.0
6.477 1047.73 164.5
7.250 935.97 184.2
8.116 836.13 206.1
9.085 746.94 230.8
10.170 667.27 258.3
11.384 596.09 289.2
12.744 532.51 323.7
14.265 475.7 362.3
15.969 424.96 405.6
17.875 379.63 454.0
20.010 339.14 508.2
22.399 302.96 568.9
25.074 270.64 636.9
28.067 241.77 712.9
31.419 215.98 798.0
35.170 192.95 893.3
39.370 172.36 1000.0
363.403inches 9230.4millimeters in total length of tubing
One side benefit is it uses almost half as much tubing....
I haven't built either one, but it makes intuitive sense to me that the resonant tubes should each share the same bandwidth, rather than be round numbers in length.
Thank you very much everybody for your suggestions. Mark, thanks for the article regarding the directional microphone and associated circuitry, which I believe will put me on the right track. Though I'm still just learning about this, it had not occurred to me to build the electrical equivalent of what the mechanical tube of a shotgun microphone is doing. Panomaniac, what you said about "the long tube having slots in it to cancel the rear waves" got me to thinking about cancellations. Then when I saw this article and circuit, I realized that this could be accomplished electrically, which could probably be more easily controlled.
The circuit itself is very simple, as the article describes. It's simply two identical electrect condensor microphone capsules (it doesn't say if they're omni or uni-directional, but I'll try both) each being amplified, then summed together out of phase into an op amp (one fed into the inverting input, the other into the non-inverting input) so that any sound common to both capsules is canceled out. That output then goes to a band-pass eq circuit that rolls off frequencies above and below the area of speech.
One needn't actually build the circuit, however. Instead, each microphone could feed it's own channel of an audio mixer, with one channel switched out-of-phase. Then when the two channels are mixed together, an eq could be engaged to roll off all but the vocal range.
The "magic" of all this is has everything to do with the physical placement of the capsules, one behind the other. The author of the article G. Baars suggests 8 inches being a good compromise for voice. Since everything common to both microphones gets canceled, it makes sense that if the sound source is located closer to the front microphone, only those sounds will get through.
I've played with out-of-phase microphones before, but it never occurred to me to apply it to a single directional microphone in this way. I've even done what the rock group the Grateful Dead used to do onstage. Perhaps you've seen videos of this. They had the PA speakers behind them, which would normally be asking for trouble. Then, they had two microphones on each stand one mounted above the other a few inches apart that were out-of-phase with each other. The singer would sing into only one of the two microphones. Everything common to both microphones (stage noise, speakers, potential feedback) would cancel out. The singer's voice, however, being UNcommon to both mics, would be the only thing that wouldn't cancel. This is a bit of what a directional cardioid microphone does mechanically anyway, but taken to another level.
Well, enough drivel. Thanks again everybody for your help!
Scott Morrison
Van Nuys, CA
The circuit itself is very simple, as the article describes. It's simply two identical electrect condensor microphone capsules (it doesn't say if they're omni or uni-directional, but I'll try both) each being amplified, then summed together out of phase into an op amp (one fed into the inverting input, the other into the non-inverting input) so that any sound common to both capsules is canceled out. That output then goes to a band-pass eq circuit that rolls off frequencies above and below the area of speech.
One needn't actually build the circuit, however. Instead, each microphone could feed it's own channel of an audio mixer, with one channel switched out-of-phase. Then when the two channels are mixed together, an eq could be engaged to roll off all but the vocal range.
The "magic" of all this is has everything to do with the physical placement of the capsules, one behind the other. The author of the article G. Baars suggests 8 inches being a good compromise for voice. Since everything common to both microphones gets canceled, it makes sense that if the sound source is located closer to the front microphone, only those sounds will get through.
I've played with out-of-phase microphones before, but it never occurred to me to apply it to a single directional microphone in this way. I've even done what the rock group the Grateful Dead used to do onstage. Perhaps you've seen videos of this. They had the PA speakers behind them, which would normally be asking for trouble. Then, they had two microphones on each stand one mounted above the other a few inches apart that were out-of-phase with each other. The singer would sing into only one of the two microphones. Everything common to both microphones (stage noise, speakers, potential feedback) would cancel out. The singer's voice, however, being UNcommon to both mics, would be the only thing that wouldn't cancel. This is a bit of what a directional cardioid microphone does mechanically anyway, but taken to another level.
Well, enough drivel. Thanks again everybody for your help!
Scott Morrison
Van Nuys, CA
scottman said:
All that stuff about two capsules is correct but very simplistic as it only applies to a single wavelength and its harmonics (multiples of wavelength).
But any simple arrangement of two microphones will create a multitude of polar patterns, totally each on wavelength - not good. That's why there are multiple holes in any modern shotgun mic - each of them functions as a mic capsule, but they cover a range of frequencies because they are many. The more holes/capsules - the wider the "directional band".
I've read about a directional mic that consists of ten or more mic capsules situated randomly (to minimize frequency peaking) on a plane and connected in parallel electrically. This way your "common to all mics" concept would be of more use. The smaller the gaps and the bigger the plane - the wider the "directional band".
k
I think what you are refering to is the comb filtering effect. It is both a potential problem and part of the solution. The distance between the mic capsules can be used to create a null in the pattern of the mic array. The slots on the pro models are a further effort to single out a useful passband for the microphone to pickup.
A shotgun mic is not wide band in the normal sense of a microphone ie: 50hz to say 15000 hz. A long range mic has a much smaller passband or frequency range that it will effectively pickup.
Hope this helps a bit.
Mark
Hmmm. Well, it appears that my cause for celebration was short-lived and premature. So, Wxn, what you're saying makes sense about the two capsules being simplistic, addressing only one wavelength and its multiples, hence the need for multiple capsules or multiple ports.
So now I'm back at square one. It would seem more cost-effective to construct a shotgun microphone using a mechanical/acoustical approach (multiple ports), as I suspect the major manufacturers are doing, rather than doing so electrically (multiple capsules).
That said, I guess it comes back to my original question about there being some general rules of thumb for constructing a shotgun microphone, such as knowing what the variables and trade-offs are and how to control them: capsule pickup pattern, tube length/diameter, number/size/placement of ports, and resultant pickup pattern/frequency response (passband). Too many variables for me to feel inspired to blindly start experimenting and essentially reinventing the wheel.
Thanks,
Scott
So now I'm back at square one. It would seem more cost-effective to construct a shotgun microphone using a mechanical/acoustical approach (multiple ports), as I suspect the major manufacturers are doing, rather than doing so electrically (multiple capsules).
That said, I guess it comes back to my original question about there being some general rules of thumb for constructing a shotgun microphone, such as knowing what the variables and trade-offs are and how to control them: capsule pickup pattern, tube length/diameter, number/size/placement of ports, and resultant pickup pattern/frequency response (passband). Too many variables for me to feel inspired to blindly start experimenting and essentially reinventing the wheel.
Thanks,
Scott
There is one more thing. Parabolic reflectors (the dish types) can be described as a high pass filter+amplifier. That old design with multiple resonators - bandpass filter, possibly with some amplification. But any tube-with-holes design is actually a directional attenuator. This means that they're only effective at short distances. Capsules' noise floor prevents this type of mic from being effectively used at greater distances. Of course the better commercial mics use high quality condenser capsules, but that only allows for more gain electrically, which does not necessarily improve overall S/N.
So what is good about multiple capsules in parallel - is that S/N goes up by 3dB with every doubling of capsules. Connecting and matching say 4 electret capsules is easy and cheap. But buying a capsule with noise floor lower by 6dB - not always. Of course one capsule mics will usually have less phasing problems and therefore more natural sound.
Going back to original question, I think that almost any rigid tube with almost any number of holes and a capsule in it will work as a shotgun mic. I've seen many commercial designs and they don't seem to follow any common pattern - I suppose every design is a result of long trial-error and guesstimation to make it work as expected.
So what is good about multiple capsules in parallel - is that S/N goes up by 3dB with every doubling of capsules. Connecting and matching say 4 electret capsules is easy and cheap. But buying a capsule with noise floor lower by 6dB - not always. Of course one capsule mics will usually have less phasing problems and therefore more natural sound.
Going back to original question, I think that almost any rigid tube with almost any number of holes and a capsule in it will work as a shotgun mic. I've seen many commercial designs and they don't seem to follow any common pattern - I suppose every design is a result of long trial-error and guesstimation to make it work as expected.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.