Hmm, you could mix 4 and 8 ohm versions of the same fullrange driver and run the middle of a three driver matrix as a bipole. If you're trying to put the thing in the middle of a room.
edit: Huh, or use resistors on the side speakers to get the impedances the same as the middle bipole, or vice versa.
Last edited:
Use a 2.1 amp, center channel to woofer, side firing left and right.
10 W / channel is enough for most rooms.
If possible, on the roof or high up near the roof.
10 W / channel is enough for most rooms.
If possible, on the roof or high up near the roof.
It strikes me that two important elements in solving this problem are separation and directivity. Separation occurs naturally with drivers to the sides, but for front facing would need some kind of protruding middle baffle to create stereo effectively. That and directivity, to compensate for the lack of conveniently placed side walls in most rooms. For omni-stereo middle of the room, you could try a Wurlitzer spinning tremolo speaker (joke).
Right, form an orderly cue to shoot this one down;
What about two 2" horn flares in "side" (M/S) with an additional "mid". A bit like this exhaustive prototype I've bit together (not). I guess it requires a 90degree corner placement (?)
Right, form an orderly cue to shoot this one down;
What about two 2" horn flares in "side" (M/S) with an additional "mid". A bit like this exhaustive prototype I've bit together (not). I guess it requires a 90degree corner placement (?)
I do like idea of a bipole solution. If the goal is big room filling sound from one box, that might be a good thing.
There's also the option of running the front and rear drivers out of phase, in other words, dipole. This gives you a figure 8 pattern, cancelling the sound that goes to the sides. That might help with separation between center and sides.
An odd artifact if you do a dipole 3-speaker solution (at least if it uses a Sum/Difference configuration) is that at the back of the speaker, left and right sides will be swapped. At least I think so, it's a bit of a mind-bender.
I don't think that would work very well.
It's essentially the same as putting two regular speakers back to back. No direct sound, only reflected (and probably asymmetrical reflections).
Wattage isn't much of an issue for the 1-box stereo challenge. It could scale from 1 to 100 watts without anything else changing much.
However, if you want to squeeze a lot of bass from a small box, you usually need a lot of watts. The Riva Stadium, for example, has 200 watts of amplification.
I presume that when you say "roof", you mean "ceiling". If it was on the roof, it would be outside, and nobody inside would hear it!
English is my second language, and I used to mix the two up as well.
Unfortunately this picture is only loading halfway. Are others having that problem?
I found that in a regular stereo pair placed near one another, an absorbent baffle helps the separation a bit.
But the problem is that as long as you sit head-on and can see both speakers, both your ears will also get direct sound from both speakers. So you get almost no directional clues from that, only from the reflected sound.
But that's essentially what the problem is for ALL 1-box stereo solutions. You need either physical separation of the speakers, or use reflected sound. There are no other options, are there?
I like the idea of a rotating Leslie speaker... If you switch the input quickly from left to right in synchronization with each rotation, you get stereo! You also get a lovely warbling doppler effect.
Put the three speakers in the same box, each speaker in its own vented enclosure, with the back of the box housing the amp.
Waiting for carpenters (long story below).
Will post sketch tomorrow.
A modified version of FM unit, crossed with a sound bar.
All the parts I have, except the plywood, which will come as scrap from a friend who uses sheets to make shipping crates, sometimes I take waste bits of plywood for use at home or work.
Waiting for carpenters (long story below).
Will post sketch tomorrow.
A modified version of FM unit, crossed with a sound bar.
All the parts I have, except the plywood, which will come as scrap from a friend who uses sheets to make shipping crates, sometimes I take waste bits of plywood for use at home or work.
Now I can see your picture.
With a full-on Sum/Difference matrix, this could work!
Using two horns kind of makes sense. By controlling the dispersion of the Side signals, you can avoid them overlapping and cancelling out (since they are the same signal, but one is inverted). A sideways dipole (open baffle or closed box) will also do that, but with the horns you have the option of pointing them forward a bit. I'd consider putting the horns on a swivel mount, so you can try to aim them at some good reflective surface.
The side signal usually contains very little bass, so the low frequency reach of the horns won't need to be very deep. But I think you do need to get down to something like 500Hz.
You probably couldn't use an "amp-hack" kind of matrix for this, since the speaker types are so different.
But it's very important to include that coffee cup in the cabinet, that's where the magic comes from!
I was imagining using 3 identical drivers. These were just to hand, if you couldn't already tell. I've experimented in the past with using 2" full range drivers in a horn designed for 2" compression drivers down to 3-400hz. Can work well with eq. If one could be found or designed with a radiation pattern that could enable the "sides" to be angled as to not overlap or interfere, with a standard non-horn mid, it could be interesting. A proper design challenge!
A 2.1 amp could work with a 4 driver set-up. The M/S matrix 3 drivers go stereo and a dedicated woofer for the bass mono.
"But that's essentially what the problem is for ALL 1-box stereo solutions. You need either physical separation of the speakers, or use reflected sound. There are no other options, are there?"
errrrr....no. ?
A 2.1 amp could work with a 4 driver set-up. The M/S matrix 3 drivers go stereo and a dedicated woofer for the bass mono.
"But that's essentially what the problem is for ALL 1-box stereo solutions. You need either physical separation of the speakers, or use reflected sound. There are no other options, are there?"
errrrr....no. ?
Gotcha, that makes sense. I'm interested in your 2" horn experiments, have you posted about it on here?
Most 2.1 amps with a built-in crossover won't run the woofer as high as is needed to reach the 3-400Hz horns. Maybe if you can find a 2.1 amp with a defeatable crossover, and use passive crossovers instead?
Thinking of it, there's a gap in the marketplace for a dedicated WAV/FAST amplifier. An inexpensive 2+2 amp with an active crossover that can go from something like 200Hz to 800Hz would be awesome.
Most 2.1 amps with a built-in crossover won't run the woofer as high as is needed to reach the 3-400Hz horns. Maybe if you can find a 2.1 amp with a defeatable crossover, and use passive crossovers instead?
Thinking of it, there's a gap in the marketplace for a dedicated WAV/FAST amplifier. An inexpensive 2+2 amp with an active crossover that can go from something like 200Hz to 800Hz would be awesome.
Last edited by a moderator:
Okay, I finished a prototype a foamcore prototype for the 6-speaker matrix posted earlier.
It's actually three separate sealed (barely!) boxes, so I can play around with angles. The tops are open, so I can play with wiring options, which is why they are wearing a plywood hat, and a vise to weigh it down.
As a reminder, the speaker layout is this:
L+ R+ R+
R- L+ L-
So in theory each box with a speaker pair is supposed to work as one unit.
The center box is the easy one, with left and right, both wired positive, so this box is, in other words, the Sum.
The left box is playing left positive as well, but joining it is the right negative signal, subtracting anything that also plays on the right channel. Presumably this blend/subtraction should happen seamlessly in the air.
The right box is the same as the left, but with the phase inverted.
This will do two things.
1. For an instrument panned hard right, a positive signal will play in the center and right hand boxes, while the negative signal in the left hand box will cancel out that instrument in the left field. It should stop the right hand signal from being heard from the left, and it should psychoacoustically push the right signal further right, to where it appears to come from outside the speaker.
2. For an instrument panned hard center, the signal plays equally in all six drivers. However, the negative signals in the side boxes will cancel out all the signal coming out of those boxes. So a center panned instrument, often a lead vocal, will only be heard from the center speaker. This should promote width and separation from the left/right.
I'm still fiddling with angles and wiring options, but for my early tests I found that I preferred to wire each side (left and right) in series (with one driver wired inverted). The original diagram uses a parallel/series configuration, which I think puts more emphasis on the center drivers.
My early results are somewhat positive. The system DOES promote the sense of width beyond the physical speakers. The left and right drives have about 14" of distance (center to center), and I measured how far apart the speakers feel, and that fell to around 32".
Not too bad, but it's not the big, room filling spacious sound I've gotten with a Mid/Side setup. Or perhaps I should say workshop filling sound, because in the workshop the Mid/Side tests worked great, but in the living room it fell apart.
Here's the downfall of the 6-speaker matrix: For it to work, you have to listen to it perfectly at ear level.
Any experienced speaker designer could probably tell you why. The two speakers in one of the boxes are as near each other as possible. Even then, one I stand up straight (rather than crouch down to ear level), the sound of the two drivers in a box now arrive at a slightly different time, noticeably breaking down the phase cancellation that this whole thing depends on. The stereo width quickly collapses.
More experiments to come!
It's actually three separate sealed (barely!) boxes, so I can play around with angles. The tops are open, so I can play with wiring options, which is why they are wearing a plywood hat, and a vise to weigh it down.
As a reminder, the speaker layout is this:
L+ R+ R+
R- L+ L-
So in theory each box with a speaker pair is supposed to work as one unit.
The center box is the easy one, with left and right, both wired positive, so this box is, in other words, the Sum.
The left box is playing left positive as well, but joining it is the right negative signal, subtracting anything that also plays on the right channel. Presumably this blend/subtraction should happen seamlessly in the air.
The right box is the same as the left, but with the phase inverted.
This will do two things.
1. For an instrument panned hard right, a positive signal will play in the center and right hand boxes, while the negative signal in the left hand box will cancel out that instrument in the left field. It should stop the right hand signal from being heard from the left, and it should psychoacoustically push the right signal further right, to where it appears to come from outside the speaker.
2. For an instrument panned hard center, the signal plays equally in all six drivers. However, the negative signals in the side boxes will cancel out all the signal coming out of those boxes. So a center panned instrument, often a lead vocal, will only be heard from the center speaker. This should promote width and separation from the left/right.
I'm still fiddling with angles and wiring options, but for my early tests I found that I preferred to wire each side (left and right) in series (with one driver wired inverted). The original diagram uses a parallel/series configuration, which I think puts more emphasis on the center drivers.
My early results are somewhat positive. The system DOES promote the sense of width beyond the physical speakers. The left and right drives have about 14" of distance (center to center), and I measured how far apart the speakers feel, and that fell to around 32".
Not too bad, but it's not the big, room filling spacious sound I've gotten with a Mid/Side setup. Or perhaps I should say workshop filling sound, because in the workshop the Mid/Side tests worked great, but in the living room it fell apart.
Here's the downfall of the 6-speaker matrix: For it to work, you have to listen to it perfectly at ear level.
Any experienced speaker designer could probably tell you why. The two speakers in one of the boxes are as near each other as possible. Even then, one I stand up straight (rather than crouch down to ear level), the sound of the two drivers in a box now arrive at a slightly different time, noticeably breaking down the phase cancellation that this whole thing depends on. The stereo width quickly collapses.
More experiments to come!
Cool! Price is about right, and the feature set looks good.
The sticking point for me, as is often the case, is that it requires a Windows PC to configure. I'm a Mac guy, and would hate to have to buy and maintain a Windows box, or configure some sort of emulator, to set it up.
Why not a smartphone app (for iPhone as well, please)?
The sticking point for me, as is often the case, is that it requires a Windows PC to configure. I'm a Mac guy, and would hate to have to buy and maintain a Windows box, or configure some sort of emulator, to set it up.
Why not a smartphone app (for iPhone as well, please)?
Last edited by a moderator:
There are a few threads on here about similar things. My experience is be prepared to eq. The other issue with what I proposed as it could start to get quite large with two full range front loaded horns. But cool. The once is my mock-up picture would be too small to cover that much frequency...
Some more tinkering...
To avoid the sitting/standing collapse of the stereo width (due to loss of intentional phase cancellation), I rotated the boxes 90 degrees.
After a few trials, I decided that the drivers are best wired (from left to right):
L+ R- R+ L+ L- R+
This way the R- and R+, and L+ and L- are next to one another, for the best phase cancellation effect.
This is quite a bit better. The actual distance between the outer drivers is 12.5", but it feels more like 40-45"!
The image is a bit more stable when moving around left to right, but as you might expect, it's nothing like a proper solid stereo soundstage.
For ease of experimenting with connections I ran them all in parallel this time. The are 6 Ohm drives, which means I was punishing the poor amp with a 1.5 Ohm load! At least I didn't play loud.
Splaying out the side speakers, or pointing them straight sideways, didn't work as well. Presumably it's because the potential for the intentional cancellation between R+ and R- (and L+ and L-) gets worse the further those drivers are apart.
A tall, flat, and wide monolith speaker box (maybe 36" high, 20" wide, 2-3" deep) with six nice 2.5" full range drivers in a row, and a flat subwoofer below, could be quite nice. Remember, with a 6-speaker matrix, any stereo or 2.1 amp will work fine.
Ultimately, summing left and right, and subtracting left from right, and vice versa, is probably still best done electronically, rather than acoustically like this.
To avoid the sitting/standing collapse of the stereo width (due to loss of intentional phase cancellation), I rotated the boxes 90 degrees.
After a few trials, I decided that the drivers are best wired (from left to right):
L+ R- R+ L+ L- R+
This way the R- and R+, and L+ and L- are next to one another, for the best phase cancellation effect.
This is quite a bit better. The actual distance between the outer drivers is 12.5", but it feels more like 40-45"!
The image is a bit more stable when moving around left to right, but as you might expect, it's nothing like a proper solid stereo soundstage.
For ease of experimenting with connections I ran them all in parallel this time. The are 6 Ohm drives, which means I was punishing the poor amp with a 1.5 Ohm load! At least I didn't play loud.
Splaying out the side speakers, or pointing them straight sideways, didn't work as well. Presumably it's because the potential for the intentional cancellation between R+ and R- (and L+ and L-) gets worse the further those drivers are apart.
A tall, flat, and wide monolith speaker box (maybe 36" high, 20" wide, 2-3" deep) with six nice 2.5" full range drivers in a row, and a flat subwoofer below, could be quite nice. Remember, with a 6-speaker matrix, any stereo or 2.1 amp will work fine.
Ultimately, summing left and right, and subtracting left from right, and vice versa, is probably still best done electronically, rather than acoustically like this.
Very good investigations. I'm not 'an experienced speaker designer' but two speakers with the same signal close together causes lobing - narrowed dispersion in the vertical axis (if they're one on top of the other) and a wider dispersion horizonally. It's often considered a positive because the wider dispersion horizontally is a good thing, and the narrowed dispersion vertically lessens near reflections. 🙂
Obviously there are other factors here!
Obviously there are other factors here!
Good points all.
Of course in this quest, especially with the 6-matrix-drivers-in-a-horizontal-line, all "proper" design considerations are out the window!
In the foam core prototype only two pairs out the 6 drivers play the same signal. But those pairs aren't very near one another, and at the high frequencies they absolutely will interfere, and mess with the dispersion pattern.
It's all a big mess, and it's all wrong. But it sounds pretty okay!
Of course in this quest, especially with the 6-matrix-drivers-in-a-horizontal-line, all "proper" design considerations are out the window!
In the foam core prototype only two pairs out the 6 drivers play the same signal. But those pairs aren't very near one another, and at the high frequencies they absolutely will interfere, and mess with the dispersion pattern.
It's all a big mess, and it's all wrong. But it sounds pretty okay!
Last edited by a moderator:
Keep our forum clean and clutter free! Please stop quoting the entire post just above yours. (see: rule 15)Over quoting has been removed.
I just discovered this old beauty - the JBL Ranger Metregon C45!
The dual louvered 2-way speakers each bounces the sound off a curved surface, presumably projecting the sound widely, and produces a very big sweet spot.
Has anyone heard one, or something like it?
The dual louvered 2-way speakers each bounces the sound off a curved surface, presumably projecting the sound widely, and produces a very big sweet spot.
Has anyone heard one, or something like it?
I'll keep this short (obvs), but thats a very cool design. Thats a compression driver right? And the curved baffle acts as an extension of the horn, which ends at the midpoint, presumably to avoid left/right overlap. I want one!
I was actually wondering if the matrix stereo was even necessary to achieve your original goal.
I was actually wondering if the matrix stereo was even necessary to achieve your original goal.