First I have a redrawing of weltersys' path length difference hand drawing. This is kind of close to scale - the horn itself is 30 feet long (as per OP's latest example), the distance from the horn mouth to the green hot tub is another 30 feet. I don't have a ruler and protractor here so some of the distances and angles will be a bit of a guess.
On the left is a single mono sub example with mains to the sides. On the right is a dual spaced sub configuration with the mains placed inside the horn. Also in the dual sub picture I have shown a straight horn on the left and a folded horn on the right.
In both of these examples I have drawn a lot of lines. In each case the line color represents the same thing.
Brown line - direct path from horn throat to zero degree on axis listening position.
Orange line - direct path from horn throat to some point off axis, maybe 20 or 30 degrees off axis depending on which sub configuration pic you are looking at.
Purple line - direct path from one of the mains to the zero degrees on axis listening position.
Blue line - direct path from one of the mains to some point off axis, maybe 20 or 30 degrees off axis.
Don't worry that some of the lines go outside the horn body - I'm not an artist and I'm not going to sketch this up into sketchup to get everything to exact scale and shape. You can figure out the intent easily enough with what I've provided.
I've also drawn a red X on the brown and orange line. This represents the path length that is eliminated by setting a 15 ms delay on the mains - so the path length that actually counts for the upcoming sims is from the red X to the point where the line meets the red line at the bottom, which is the horizontal axis of the hot tub listening position. All sims will reference points along the horizontal red line.
I've also drawn some short black tic marks intersecting the red horizontal line. These are approximate mic locations for the 0 degree, 5.71 degree, 11.31 degree and 21.8 degree measurement points that are common to all the following sims.
An externally hosted image should be here but it was not working when we last tested it.
On the left is a single mono sub example with mains to the sides. On the right is a dual spaced sub configuration with the mains placed inside the horn. Also in the dual sub picture I have shown a straight horn on the left and a folded horn on the right.
In both of these examples I have drawn a lot of lines. In each case the line color represents the same thing.
Brown line - direct path from horn throat to zero degree on axis listening position.
Orange line - direct path from horn throat to some point off axis, maybe 20 or 30 degrees off axis depending on which sub configuration pic you are looking at.
Purple line - direct path from one of the mains to the zero degrees on axis listening position.
Blue line - direct path from one of the mains to some point off axis, maybe 20 or 30 degrees off axis.
Don't worry that some of the lines go outside the horn body - I'm not an artist and I'm not going to sketch this up into sketchup to get everything to exact scale and shape. You can figure out the intent easily enough with what I've provided.
I've also drawn a red X on the brown and orange line. This represents the path length that is eliminated by setting a 15 ms delay on the mains - so the path length that actually counts for the upcoming sims is from the red X to the point where the line meets the red line at the bottom, which is the horizontal axis of the hot tub listening position. All sims will reference points along the horizontal red line.
I've also drawn some short black tic marks intersecting the red horizontal line. These are approximate mic locations for the 0 degree, 5.71 degree, 11.31 degree and 21.8 degree measurement points that are common to all the following sims.
Last edited:
The first issue I will address is straight sub horn vs folded. Weltersys has advocated for a folded sub several times (and I did too at the beginning, but my advice included a LOT of modular units, not a single huge folded horn or even as few as 4 folded sub horns).
1. A folded horn as shown will give two parallel walls all the way down the length of the horn. This is a huge mistake. It is possible to eliminate the parallel walls but that would make the design and construction extremely complex.
2. It's very difficult to fold a T=1 pure exponential flare into anything resembling a rectangular shape because it flares out so fast near the mouth. And as shown, unless you make a lot of 180 degree bends you are going to have a lot of wasted space in the "box".
3. It's nearly impossible to make a folded horn end loaded. Assuming it has to be Od, the throat would have to be 12 feet high and have to have an extremely high aspect ratio to accomodate the eight 18 inch drivers.
4. It's very difficult to make a folded horn interchangable from one 12 inch driver to eight 18 inch drivers. I suppose it's possible but it certainly wouldn't be a rectangular folded horn if designed for that purpose.
4. In the dual sub configuration, if you look at the brown line length compared to the purple line length (from the red X to the horizontal red line at the bottom) and also compare the length of the orange and blue lines, both of these colored pairs are almost exactly the same length. So the straight horn path length problem weltersys was complaining about simply does not exist. Since the path length problem does not exist there's no need to "fix" it by folding the horn.
5. OP wants a straight horn. As if the first 4 points here weren't enough to prove a folded horn is not necessary or desirable, this point overrides everything and dictates that the horn WILL be straight so there's not even that much point in debating it.
1. A folded horn as shown will give two parallel walls all the way down the length of the horn. This is a huge mistake. It is possible to eliminate the parallel walls but that would make the design and construction extremely complex.
2. It's very difficult to fold a T=1 pure exponential flare into anything resembling a rectangular shape because it flares out so fast near the mouth. And as shown, unless you make a lot of 180 degree bends you are going to have a lot of wasted space in the "box".
3. It's nearly impossible to make a folded horn end loaded. Assuming it has to be Od, the throat would have to be 12 feet high and have to have an extremely high aspect ratio to accomodate the eight 18 inch drivers.
4. It's very difficult to make a folded horn interchangable from one 12 inch driver to eight 18 inch drivers. I suppose it's possible but it certainly wouldn't be a rectangular folded horn if designed for that purpose.
4. In the dual sub configuration, if you look at the brown line length compared to the purple line length (from the red X to the horizontal red line at the bottom) and also compare the length of the orange and blue lines, both of these colored pairs are almost exactly the same length. So the straight horn path length problem weltersys was complaining about simply does not exist. Since the path length problem does not exist there's no need to "fix" it by folding the horn.
5. OP wants a straight horn. As if the first 4 points here weren't enough to prove a folded horn is not necessary or desirable, this point overrides everything and dictates that the horn WILL be straight so there's not even that much point in debating it.
Last edited:
Next I'm going to tackle the comb filtering issue with the single mono sub vs stereo separted subs. Both examples have mains included this time.
As i have mentioned already too many times now, at mid and high frequencies the combing notches are pretty narrow, 1/2 octave or less. This will look horrific in a sim but will not impact your listening pleasure much. But at frequencies below about 200 hz you get deep notches that are 2 octaves or more wide at certain locations. This is a horrific mess, it's definitely audible and it will definitely impact your listening pleasre.
For these examples the measurements are as follows -
Single mono sub placed on zero degree listening axis, 30 feet from the horizontal red line where all the virtual mic positions are located. Mains are placed with 8 meter center to center distance, so each is 4 meters off the center line.
Dual separated subs are placed pointing at the zero degree listening axis but are placed with an 18 meter center to center separattion, so each is 9 meters off the centerline. The mains are inside the subs, exactly centered, so they also have the same 18 meter center to center distance and are each 9 meters off the center line.
I also adjusted the depth position variable - in the single sub example the brown vs purple and orange vs blue lines are different lengths, so at different mic positions they are not only not on the same horizontal axis, but also not on the same Z (depth) axis. At the different mic locations the position of the mains in relation to the sub changes by about 3 meters, at the zero axis position the subs are about 2 meters in front of the mains, at 11.31 degrees they are ~ equal distance and at 20.8 degrees the mains are about 1 meter in front of the sub on the Z axis. This is all accounted for in the sim for each virtual mic location. But in the dual sub configuration, the brown vs purple and orange vs blue line lengths are all close enough to zero path length difference that I didn't have to adjust the Z axis for any of the dual sub mic locations.
As i have mentioned already too many times now, at mid and high frequencies the combing notches are pretty narrow, 1/2 octave or less. This will look horrific in a sim but will not impact your listening pleasure much. But at frequencies below about 200 hz you get deep notches that are 2 octaves or more wide at certain locations. This is a horrific mess, it's definitely audible and it will definitely impact your listening pleasre.
For these examples the measurements are as follows -
Single mono sub placed on zero degree listening axis, 30 feet from the horizontal red line where all the virtual mic positions are located. Mains are placed with 8 meter center to center distance, so each is 4 meters off the center line.
Dual separated subs are placed pointing at the zero degree listening axis but are placed with an 18 meter center to center separattion, so each is 9 meters off the centerline. The mains are inside the subs, exactly centered, so they also have the same 18 meter center to center distance and are each 9 meters off the center line.
I also adjusted the depth position variable - in the single sub example the brown vs purple and orange vs blue lines are different lengths, so at different mic positions they are not only not on the same horizontal axis, but also not on the same Z (depth) axis. At the different mic locations the position of the mains in relation to the sub changes by about 3 meters, at the zero axis position the subs are about 2 meters in front of the mains, at 11.31 degrees they are ~ equal distance and at 20.8 degrees the mains are about 1 meter in front of the sub on the Z axis. This is all accounted for in the sim for each virtual mic location. But in the dual sub configuration, the brown vs purple and orange vs blue line lengths are all close enough to zero path length difference that I didn't have to adjust the Z axis for any of the dual sub mic locations.
So let's get to the pretty pictures.
First pic - this is just mains playing fullrange shown at 0, 2.86, 5.71. 8.53, 11.31, 21.8, 26.57, 33.02, 38.66 and 45 degrees. It looks nasty but for the most part those notches are not very wide, so pretty benign. The exception is the lowest one, it's two octaves wide and moves down in frequency as you move off axis.
Second pic - this is the single sub configuration with mains added. Crossover is 80 hz 4th order LR for both the mains (high pass) and the subs (low pass). The 4 steps show 0, 5.71, 11.31 and 20.8 degrees, roughly corresponding to the black tic mark virtual mic locations on the picture. As you can clearly see, at 0 degrees there's already a problem, a 7 db notch at the crossover. At 5.71 degrees you get a 2 ocatave wide notch at 200 hz and as things move further off axis there's no more deep 2 octave side notches. In fact, below the crossover point this system is very well behaved at all points on and off axis.
Third pic - this is the dual sub with mains inside the subs. Crossover is the same, 80 hz 4th order LR for both subs and mains. Clearly, on axis everything is beautiful. At 5.71 degrees there's a massive 2 octave wide notch at 100 hz and as you go futher off axis the massive notch lowers in frequency.
Fourth pic - this the dual sub with mains configuration with a different crossover. I've moved it up to 300 hz LR for both the subs and mains. Clearly this isn't any better (and not a whole lot worse) than the same configuration with a lower crossover.
Hopefully these gifs will all play properly. In case it's not clear, in each case the flattest summed response (black line) is the 0 degree start of the gif. It should also be clear by looking at the red and blue lines which show the mains and subs individual contribution.
First pic - this is just mains playing fullrange shown at 0, 2.86, 5.71. 8.53, 11.31, 21.8, 26.57, 33.02, 38.66 and 45 degrees. It looks nasty but for the most part those notches are not very wide, so pretty benign. The exception is the lowest one, it's two octaves wide and moves down in frequency as you move off axis.
Second pic - this is the single sub configuration with mains added. Crossover is 80 hz 4th order LR for both the mains (high pass) and the subs (low pass). The 4 steps show 0, 5.71, 11.31 and 20.8 degrees, roughly corresponding to the black tic mark virtual mic locations on the picture. As you can clearly see, at 0 degrees there's already a problem, a 7 db notch at the crossover. At 5.71 degrees you get a 2 ocatave wide notch at 200 hz and as things move further off axis there's no more deep 2 octave side notches. In fact, below the crossover point this system is very well behaved at all points on and off axis.
Third pic - this is the dual sub with mains inside the subs. Crossover is the same, 80 hz 4th order LR for both subs and mains. Clearly, on axis everything is beautiful. At 5.71 degrees there's a massive 2 octave wide notch at 100 hz and as you go futher off axis the massive notch lowers in frequency.
Fourth pic - this the dual sub with mains configuration with a different crossover. I've moved it up to 300 hz LR for both the subs and mains. Clearly this isn't any better (and not a whole lot worse) than the same configuration with a lower crossover.
Hopefully these gifs will all play properly. In case it's not clear, in each case the flattest summed response (black line) is the 0 degree start of the gif. It should also be clear by looking at the red and blue lines which show the mains and subs individual contribution.
The big issue here is not the comb filtering at mid and high frequencies, those notches are not very wide and won't be distracting. The big issue is the lower frequency notches that are deep and two octaves wide, these are a huge problem.
The single sub configuration has only one such notch, it's at 200 hz and only appears at 5.71 degrees off axis. The dual sub configuration has this deep notch at EVERY off axis position at some frequency.
If you average the results, the single sub comb filtering as a whole is a lot more calm than the dual sub configuration. Furthermore, when you chop that single horn up into modular pieces, all the pieces will comb filter with each other (this is not shown here, maybe I'll sim that up later).
What I cannot show in sims is weltersys' insistance that the soundstage will wander around at the crossover point with the single sub setup. All logic and empirical evidence shows this is not true, especially not with a steep 80 hz crossover, but I can't prove it either way.
What I can prove is that the comb filtering is a lot worse with the dual subs than the single sub and I think I've done that pretty effectively here.
No matter what you do there's going to be a lot of comb filtering, but the point is that you can minimize it's effect by eliminating most of the instances where it's causing a deep 2 octave wide hole.
The single sub configuration has only one such notch, it's at 200 hz and only appears at 5.71 degrees off axis. The dual sub configuration has this deep notch at EVERY off axis position at some frequency.
If you average the results, the single sub comb filtering as a whole is a lot more calm than the dual sub configuration. Furthermore, when you chop that single horn up into modular pieces, all the pieces will comb filter with each other (this is not shown here, maybe I'll sim that up later).
What I cannot show in sims is weltersys' insistance that the soundstage will wander around at the crossover point with the single sub setup. All logic and empirical evidence shows this is not true, especially not with a steep 80 hz crossover, but I can't prove it either way.
What I can prove is that the comb filtering is a lot worse with the dual subs than the single sub and I think I've done that pretty effectively here.
No matter what you do there's going to be a lot of comb filtering, but the point is that you can minimize it's effect by eliminating most of the instances where it's causing a deep 2 octave wide hole.
Last edited:
Hi Entropy455,
Post # : "...I'm going to build a small gantry crane over the rear of the horns..."
Mission creep, or, a man's gotta do what a man's gotta do? 🙂
Regards,
Post # : "...I'm going to build a small gantry crane over the rear of the horns..."
Mission creep, or, a man's gotta do what a man's gotta do? 🙂
Regards,
Eric,
A troll has been repeating his rant that dual subs will have severe comb filtering and "cause considerable sonic disadvantage", while he also knows very well the same applies at all frequencies- stereo mains will have exactly the same "considerable sonic disadvantage" in the frequency ranges our hearing is progressively more sensitive to.
The troll endlessly repeats himself saying that the comb filtering of dual subs, which cover only 2 or 3 of the ten octaves of our hearing, octaves that are arguably the least musically important, easily verified by the fact that most people seem happy to listen to TV, radio and computers with virtually no response in the three octaves below 160 Hz.
I have still not encountered even one "bass head" happy to listen a sub without a top speaker playing above 100 Hz.
Somehow in the alternate troll universe, bass comb filtering is somehow worse than the comb filtering we hear anywhere but dead center that is created in the coverage pattern of any top cabinets with shared signal (shared signal meaning any audio source that is not hard-panned completely left or right) in a stereo system.
Enough with that nonsense, lets get to your plans and questions:
1) A "proper exponential expansion" can be done with three non-curved walls and will measure and sound virtually the same as bass horn with four curved walls, or a circular curved expansion. I used the caveat "bass" as upper frequencies coverage pattern will be affected by the horn shape, but if the bass horn is not used high, the shape is immaterial (other than the concrete and iron material used).
2) "Choking" a huge horn mouth is not a problem, you have probably attended plenty of concerts and nightclubs where people are nearly covering the bass horn mouth, the bass waves diffract around those objects like water through a sieve.
A mains speaker cabinet is different than "meat puppets", it is generally made of hard, reflective materials that will reflect a portion of the sound back into the horn, the return of which will have "out of phase" (time) components that will muck up the upper bass sound.
3)The "old engineering texts" generally only say a horn can be good for a decade of reproduction, upper frequency being 10x lowest, 15 Hz to 150 Hz.
To some extent, that limitation was based on the so called "Doppler Distortion" common with low-excursion drivers. When drivers are pushed beyond Xmax, the HF distortion compliment is modulated by the frequency of the lower tones.
"Doppler Distortion" does not exist in the true sense, the frequency shift caused by cone movement is far to little to be heard (it would be on the order of one millionth of a musical note, or at worst maybe a small fraction of the "temper" built into piano tuning to make them "sound right"), but A.M.(amplitude modulation) distortion is a very real phenomenon, and sounds horrible.
Take a 20 Hz tone to your bass set up, turn it up so the cone is moving beyond Xmax, (probably only a few watts needed below Fb) then play a vocal track through it at the same time- the vocal track will sound "garbled" until the tone is reduced below Xmax.
As long as you don't play your subs above Xmax/Xvar, A.M. ("Doppler") distortion is of no concern.
4) From a frequency response or fidelity standpoint there is no reason a well built straight or folded bass horn can not sound excellent from 15 to 150 Hz, but from one important sonic aspect, I'd recommend not crossing to the upper cabinets above 100 Hz.
You have stated your bass horns will be on the ground.
Because of #2, and to mitigate the Inverse distance law, you will likely want the top cabinets elevated at least to the top of the bass horns, or perhaps flown as high as a "telephone pole".
The LF fundamental notes of instruments will appear to "walk" between the two cabinets if the crossover frequency is too high. At 100 Hz I hear no "walking", at 160 Hz I hear the effect easily when using 24 dB per octave crossovers. I would not use less than a 24 dB per octave crossover for subs on the ground, people standing in front of them "muddy" the sound above 100 Hz.
5)Designing for an octave of overlap is a good "quality assurance" that the system will sound good through the crossover overlap range, which even with 24 dB per octave crossovers covers around an octave. The actual response of my 1979 dampened fiberglass straight exponential horns are shown below, old-timers will notice a "chart recorder" was used.
As you can see by the wiggles, I did not rest my finger on the pen in order to "smooth" the response like manufactures often did in the "olden days".
Now all that needs to be done to achieve a smooth measured frequency response is hit the "smoothing" tab on a measurement program screen, and an almost straight line appears.
Also note the "1 watt at 4 feet", it was some time later that "one meter" measurements were common in the USA- Jimmy Carter was unsuccessful in getting the metric system adopted here, because folks are afraid of something "new" even though it may be far superior.
The single 45" x 45" x 30" deep "bass" horn raw (unequalized) low response was "happening" to 60 Hz, with the "full stack" as pictured, with four bass cabinets per side it did 40 Hz no problem. With a single cabinet well over 107 dB one watt one meter, the on axis forward sensitivity of four cabinets was around 98 dB one watt one meter at 40 Hz, with about 3200 peak watts for the 8 drivers, that works out to about 122 dB.
One Keystone sub using a single B&C 18SW115-4 can do an easy 128 dB at 40 Hz, and is less than 1/8th the size of just the bass cabinets in that old system. Times changed, I had to retire the straight horns.
The response of the 45" x 22.5" x 24" deep low mid horn rolls off below 160 Hz, it is about -10 dB at 90 Hz, I used the cabinet for "bass" on some lightweight gigs, a simple boost on the graphic EQ was all that was required to satisfy the kick/bass quotient. We crossed between the 15" and the 12" at 200 Hz, but (almost) always stacked on cases, stages, or scaffold to get the cones at least head height for the reasons explained in #2 and #4.
6) As the individual "point sources" of an orchestra are all discreet (no shared outputs) sources of sound, no lobing or "comb-filtering is possible.
That said, large instrument "point sources" such as the head of a kettle drum or harp of a piano do create both lobing and comb-filtering, which is part of the sound we hear "live". Of note, a close-miced piano lacks the lobing and comb-filtering we hear playing it or sitting in a concert hall, and many find it to sound "dry" or "lifeless".
Cheers,
Art
Attachments
Last edited:
As you can very clearly see, in the mid and high frequencies the comb filters are VERY narrow and won't even show up in a measurement except for a slightly rippled response. They probably won't show up at all in an overly smoothed measurement. The comb filtering at mid and high frequencies does not sound bad (it is noticeable but doesn't sound bad), I already said this and said that if it did sound bad no one would listen to stereo much less 7.1 channels. You are very deliberately misrepresenting what I've said a number of times.
At the lower frequencies where the notches are TWENTY DB DEEP AND TWO OCTAVES WIDE this will be clearly audible and measurable, notches like this don't smooth out like very narrow notches at higher frequencies do. You would have to be completely deaf to miss a very deep 2 octave wide notch, and your constant attempts to downplay it don't make any sense and don't lessen the importance of the issue.
Repeating this doesn't make me a troll and ignoring it doesn't do much for your credibility.
OMG, this is exactly what I've been saying for the last 20 pages. You kept arguing against it and calling me a repeating troll and now you agree with me. Will wonders never cease.
I said very clearly with a low and steep crossover there would be no wandering soundstage.
Finally, flying the mains on top of a telephone pole as you are now advocating is not necessary or desirable. It won't do anything good, but it will introduce comb filtering in the 3rd dimension that hasn't been discussed yet, vertical combing. Usually vertical combing isn't a big deal, just steer the lobe out of the way. But in this case, with such a HUGE center to center distance and with the mains placed so high and pointing down the combing will be as much of a problem as the combing in the vertical and depth axis. What you are advocating is literally going to be a checkerboard pattern of massive peaks and nulls all through crossover overlap frequencies across the entire defined audience area.
At this point I'm not even sure if you believe what you are saying or if you are doubling down on your past terrible recommendations to save face by hoping that most people are not bright enough to realize how bad what you are recommending will be but either way it doesn't matter. Your recommendations are a checkerboard wasteland of massive peaks and nulls two octaves wide.
Last edited:
Dear Troll,
The OP, Eric, suggested the use of telephone poles to lift his mains rather than the crank lifts I suggested. Whatever gets the speakers safely in the air and pointed down is OK by me, though obviously nobody cares about my free advice anyway, as everybody knows "you get what you pay for" 😀.
If you were not a troll I'd bother to show you 20 dB measured octave wide holes in response of mid high speakers due to comb filtering.
As far as deaf people not hearing something that is -20dB down from the main response, that is a good one to repeat, it is hard to hear something perceived as less than 1/4 as loud, you go troll!
I have another bit for you to repeat: "You would have to be completely blind to not see the black hole you live in" 😉 .
Cheers,
Art
Last edited:
He was talking about lifting them a bit, you are talking about lifting them 30 feet in the air. There's no need to raise them more than about 6 feet, that will get the sound over the crowd and will cause minimal problems. At 30 feet height you are just asking for a checkerboard pattern of huge peaks and nulls all around the crossover overlap region.
While I have no doubt that you have trouble setting up even a simple pair of stereo speakers, this is certainly not the norm. Comb filtering in the mids and highs are usually not deep notches and not usually anywhere near an octave wide.
Amusing that you don't seem to mind completely missing notes over two whole octaves but irrelevant, most people will have a huge problem with that. You continue with the personal attacks instead of presenting technical information. This is a classic sign of a weak argument - any credibility you used to have is crumbling pretty fast.
I am wondering if it is appropriate to model my bass horns as a point source. In example 1, I've got two 18" drivers spaced 15 feet apart, playing a bass note (not a bad sketch. . .) Bass wavelengths are large compared to the relatively small driver, thus the sound will radiate out from the "point" source.
In example two, the radiated source is not a point, but rather a 15' x 15' horn mouth area. The bass note "wave" is already well developed at the source (15 feet in diameter).
In comparing the two sketches: the angle-of-inclination between the wave fronts in Example 1 is quite large. There will be appreciable destructive interference (aka lobing). In Example 2, the angle-of-inclination is much smaller - where constructive interference (aka signal coupling) is likely to occur. This can also be shown mathematically, using a lot of trig functions - but the sketch works too.
Your own drawing shows the waves of the two horns intersecting. Does that answer your question?
The only thing I can't properly sim (at all) is the directivity and beaming of the horns but that doesn't matter a whole lot within 60 degrees or so and especially at frequencies 80 hz and lower.
This isn't special (doesn't break the rules of physics) just because it's a huge subwoofer. How often do you see tweeter horns side by side? Not very often and there's a good reason for that. The only time you see subwoofers side by side is when they are very narrow and/or used with a very low crossover or have been set up by someone that doesn't know what they are doing.
Last edited:
I wasn't really asking a question. My sketch shows (pictorially) what the Mutual Coupling equation on post #802 shows mathematically.
Signal coupling (aka constructive interference) happens when sources are within 1/4 wavelength. Graphically, this represents a low angle-of-inclination of the combining waves. The coupling phenomena is essentially gone when the sources are at 1/2 wavelength (which is a greater angle-of-inclination).
I don't think the Mutual Coupling equation is valid when comparing the large mouth of a bass horn - as it is far from being a localized source. When I place two bass horns side-by-side, the waves are introduced into the space side-by-side - and specifically not 15 feet apart.
Eric,
The sketch works perfectly at one frequency, but music contains an infinite range of audible frequencies.
In terms of wavelength, the boundaries of a horn may either define pattern control, or define the frequency where the horn is too small to control the pattern of a long wave.
As Ivan Beaver usually says, the answer to your wondering is "it depends".
You are contemplating a large scale installation in which the sub-harmonic frequency is within 1 Hz range of the earth's "Schumann resonance", which happens to correspond closely to ULF submarine communication frequencies.
Perhaps you are familiar with the complaints of the people of Wisconsin regarding the ULF radiation stretching across many miles of countryside used for communication with USA submarines, perhaps that may give you something in your wheelhouse you can relate to regarding your questions.
But again, perhaps my response is too "cryptic", sorry, just trying to inform those that may be interested .
Good night,
Art
Attachments
Most of us work in rooms, where destructive interference is a necessary and desirable part of the multi sub approach.
Like sands through the hourglass.. oh, the irony.
Here's what Direct says about 2 large subs with a 16 foot center to center spacing, as your new design contemplates (assuming it's a side by side dual horn, not four horn mouths as you were contemplating yesterday or the day before). Hopefully Direct has some directivity built in, but still it won't be as much as your horns.
Shown at 40, 80 and 160 hz.
Not as good as a single central mono sub but much better than dual subs with 60 foot spacing.
In the past couple of months I've been able to convince you of only a couple of things and only after a hard fought battle. When you get an idea of something you want you hold onto it for dear life until it's proven beyond any doubt that it's not the best approach, and sometimes you like to hold on even after you have that proof.
So if you want to build dual side by side horns go right ahead. I've proven the superiority of the single mono central sub approach in a few different ways now and even weltersys is agreeing that the soundstage won't wander if you use a low steep crossover.
I've mentioned several times that at sub frequencies there is little recorded media in stereo, there is no stereo effect due to inability to localize subs and the subs aren't far enough apart to perceive a stereo signal from a distance anyway. So you have the technical superiority of a single central sub and three good reasons not to bother even trying to make your sub play stereo.
But if you want stereo subs go ahead. I'm getting a bit weary defending my recommendations against the constant barrage of questioning and personal attacks and I've got no extra evidence or proof that I can use to convince you. Maybe tomorrow I'll think up yet another way to prove my technical comments but for today the tank is empty. So if you really want stereo subs despite the problems it will cause and despite the fact that you won't hear anything in stereo below 80 hz anyway, go ahead and do it. I don't care.
Shown at 40, 80 and 160 hz.
An externally hosted image should be here but it was not working when we last tested it.
Not as good as a single central mono sub but much better than dual subs with 60 foot spacing.
In the past couple of months I've been able to convince you of only a couple of things and only after a hard fought battle. When you get an idea of something you want you hold onto it for dear life until it's proven beyond any doubt that it's not the best approach, and sometimes you like to hold on even after you have that proof.
So if you want to build dual side by side horns go right ahead. I've proven the superiority of the single mono central sub approach in a few different ways now and even weltersys is agreeing that the soundstage won't wander if you use a low steep crossover.
I've mentioned several times that at sub frequencies there is little recorded media in stereo, there is no stereo effect due to inability to localize subs and the subs aren't far enough apart to perceive a stereo signal from a distance anyway. So you have the technical superiority of a single central sub and three good reasons not to bother even trying to make your sub play stereo.
But if you want stereo subs go ahead. I'm getting a bit weary defending my recommendations against the constant barrage of questioning and personal attacks and I've got no extra evidence or proof that I can use to convince you. Maybe tomorrow I'll think up yet another way to prove my technical comments but for today the tank is empty. So if you really want stereo subs despite the problems it will cause and despite the fact that you won't hear anything in stereo below 80 hz anyway, go ahead and do it. I don't care.
Of course, but how is an in room multi sub setup even remotely applicable to this discussion? Even in rooms the mid/high frequency comb filtering shouldn't be deep notches an octave wide, which was the point you were quoting when you started talking about multi subs in rooms.
When have you ever seen me go on a namecalling rampage posting absolutely nothing technical and writing a short personal attack essay?
If you have a problem with the TECHNICAL information I present please bring it up and I'll prove what I'm saying as I always do. As far as posts like this, I'm not interested in your opinion but as always I will defend myself.
For clarification - destructive and constructive interference in audio (regarding sound wave propagation) does not literally mean destroying the sound waves, or combining the sound waves into one.
Assume you are floating in a large lake. Then two boats drive by - one on the left, and one on the right - both at some arbitrary speed & distance. When the two wakes meet up with you - at any particular moment in time during the event, they might combine to lift you well above the normal surface waterline. They might combine to lower you well below the normal surface waterline. They might combine to have no change whatsoever for your position on the waterline. The wakes (waves) are not really ever truly combined or destroyed - they pass freely through each other. The important variable is what happens to you, at your particular position during the event. You'll either get a lift (constructive interference), or you'll get nothing (destructive interference). That's my mechanical explanation anyway. .. .