Hi Guys
Tomelex, you have a good ear for details, you are correct, to the right side about three houses down IS a right angle bend in the road. It was me that walked out to the road, I was trying to be quiet haha. It was a very quiet day, only a few little noises like somebody hammering on a different street etc.
Steve, I can ‘t tell you guys exactly how it was done but I can tell you and Fas42 the process.
Well, I am not sure where to start.
I have been interested in sound most of my life, I am an inventor, not a formally trained engineer.
About 30 years ago, I was lucky to have worked for a company and boss that purchased one of the first TEF machines, a TEF 10.
It (at the time) was amazing in that one could measure things in acoustics rapidly that could not be measured at all or at least very slowly. While my job was developing acoustic levitation transducers for space based container less manufacturing, loudspeakers were my interest.
Over the following years I got to use the newer TEF machines etc at Intersonics, we became known for “goofy / exotic acoustic measurements”. One experience was pretty weird / fun / interesting for me, if you like Egypt;
http://www.livesoundint.com/archives/2000/julyaug/pyramid/pyramid.php
A continuously troubling aspect was in tying what one saw in a measurement to what one hears as a sonic property.
Anyway, my point is that after doing a million measurements on all kinds of stuff, you start to assemble your own view of “how things must work”.
Sound is one of the few things that is near black art, I mean the principals are known but what one has in a given real situation is often very hard to discern being in 3d , invisible and frequency variable.
Coherent addition;
Take a subwoofer at a given drive level. Measure the amplitude AND polar distribution. It have a response and having no directivity (being small compared to the wavelength) has a spherical radiation pattern (or half when on the ground).
Take a second identical subwoofer, same drive level, place it very close to the first and now one has coherent addition and also still has spherical radiation. The level is raised 6 dB however not the 3dB one would expect by simply doubling the power. The difference is when the sources are acoustically small and less than about ¼ wavelength apart, they “feel” each others radiation resistance, the y sum into a single larger more efficient source.
If one were to reverse one of the two, in a perfect world, they totally cancel out, the principal behind active sound cancelation.
Interference pattern;
Take the two subwoofers, in phase again, place them farther apart at say 1 / 2 wavelength at frequency X. Now one finds an interference pattern, cancelation nulls are crated in line with the sources and a doughnut radiation pattern is formed. At spacing’s larger than about 1/3 wavelength, one produces an interference pattern of lobes and nulls. The larger the acoustic spacing between sources, the larger the number of lobes and nulls in the interference pattern. Also, now if you were to reverse one source, you simple re-arrange the radiation balloon.
Consider what is required IF one were designing a speaker one of who’s goals is to provide the same frequency response and phase over a large listening area? I don’t mean the two hot seats at the couch but in a movie theater, music venue or Church? It is EXACTLY the things like the interference between drivers that causes the lobes and nulls, things that cause one seat to measure much differently than another even when the room is excluded, it is a homogeneity in the pattern one is trying for, ultimately, to produce a portion of a simple expanding sphere over the whole band.
Consider what is required if you were designing a speaker to be used in a large reverberant room?
The primary object here is to confine the radiation pattern, to produce the highest “front to back” ratio so that the sound goes where the people are and as little as possible anywhere else.
A strong part of what governs a speakers polar or spherical radiation patterns is the interference between drivers that produces lobes and nulls instead of a simple pattern.
Consider what is required if you were designing a speaker who’s spectral response should not change with distance only the loudness, it must be constant directivity.
Anyway, in order to make speakers that did this, I needed to eliminate the interference between drivers.
In the 12 years or so I have been working on these single source speakers, starting with the Unity horns and current Synergy horn, there has been a progression towards an actual single acoustic source to where now an SH-50 can reproduce a square wave over a decade wide, spanning the crossovers which have no apparent phase shift. As these got better and better, they had several sonic properties which were not obvious in any measurements. If I listened to just one speaker I heard two things, the sound got “simpler”, less busy somehow (this I assumed was because the time smear was getting smaller and smaller) and the weird one, it got harder to hear how far away it was with your eyes closed.
I mean standing say 10 feet away with your eyes closed, listening to a voice, one can easily hear the exact direction, BUT it became very hard to hear how far away the speaker was. A speaker like the SH-50 for example, one can remove the grill and walk up to it and even with your head in the horn mouth, the source is floating somewhere in front of you.
I concluded that one can hear things” about a radiation pattern such that one can obviously hear the source, all the way to real no clue how far away it is. If the radiation pattern is a simple balloon, there is much less information about “where it is” and so what you hear is more of the unfettered source. While our products are rarely used to create a stereo image in commercial applications, the best part for me was that in a living room, the huge forward directivity and lack of source signature produces a totally real phantom mono image or an image anywhere between the two speakers and the enormous headroom made them effortless at any level I would choose.
If you have ever recorded or done live sound, you know her is no way around multiple mics and yet, the more you use, the less real it sounds. A problem similar to the one with multiple sources exists, if you have more than one microphone, you have encoded an interference pattern. You brain interprets these as spatial information however, the clues are mixed up relative to how you would hear them and so even two mic recordings are a pale image compared to live. In the case of the recording thingy, by not corrupting that information, you brain is free to interpret the incoming as depth and such or in the case of tomelex, could hear (correctly) a bend in the road several hundred feet away. That stuff was simply preserved without an interference pattern so your brain could recognize the patterns more firmly.
Where the loudspeakers can supply a continuous image in a full circle with 5-6 channels, I wanted a way to capture that sonic image live. Ultimately (a dream from the way old days) I want to capture and reproduce the upper half of our hearing sphere. One has to collect directional information but not encode the interference pattern multiple microphones collect.
Lastly, to be clear, I have put down what I see as of now, science is provisional, subject to change given a better explanation so things may be different tomorrow. On the other hand, we don’t even sell to the home market so I figure I can say what I think . I am glad there is some interest, I love this stuff.
If you’re interested there are a couple other recordings that might be fun, probably wait until your SAF if so equipped is out of the house before the train and be careful of the fireworks this can tax the most stud like stereo.
Bottom of the page here;
http://www.danleysoundlabs.com/technical downloads.html
Best
Tom Danley
Danley Sound Labs