Angled array with two drivers

[Squeak]

I'm looking for ways to decrease the tendency of larger FR speakers to beam HF sound. Or in other words to extent the spread of HF sound from +5 drivers.
I've done some experiments with vertical slots and balls in front of the driver, all with very little positive effect and a whole lot of negative in the form of loss of detail and colouration.
I've tried pointing the drivers upwards, to reflect off the ceiling, which sounds interesting, but looses an unacceptable amount of detail. I've tried diffuser cones with similar, if not worse results.

I love the sound of FR drivers in the sweet spot, but sometimes it would be nice to be able to change your position in the couch or walk around the room without loosing the upper detail.

I know about line arrays where multiple vertical mounted drivers gives you greater horizontal dispersion, but at the cost of interference at higher frequencies because of lope interaction.

Then it struck me: To fix the directivity problem of single FR drivers, why not have two drivers, one mounted above the other, one pointed slightly to the left and one slightly to the right? Together the two drivers would have quite a large angle of horizontal HF dispersion, with negligible HF interaction between the HF lopes.

Would this work? Has it been done before? Or would two drivers just pointing in exactly the same direction give similar or better results?

[adason]

Originally Posted by Squeak

why not have two drivers, one mounted above the other, one pointed slightly to the left and one slightly to the right?

good idea, but I would place them horizontaly...

[Squeak]

Why?
General speaker building wisdom tells us that it's never a good idea to place two drivers covering the same range next to each other horizontally.
and it would destroy the, albeit small, horizontal spread effect on the mid frequencies, that a small "array" has.

[adason]

well, I tried that once with two Lowthers and it sounded bad, don't really know why
maybe because our ears are arranged horizontaly on our head and not one above the other
it all depends how far you angle them, if the angle is small, something might happen in midrange area, if its wide angle, you are approching direct/reflected type of arrangement
totaly different thing
I think you should experiment and repot back

[Squeak]

I was thinking a small angle, so the two HF lobes just slightly overlap (I know they are very diffuse but still) and together combine into a lobe with larger horisontal angle. Polar maps for the FE126 drivers I'm using would come in handy I think... Anyone?

[David Pinnegar]

Hi!

Lowther got over the problem in designs using fiercely agressive treble with huge magnets pointing back at the wall at 45 degrees reflecting the sound off the wall - the dual position acousta. A PM2 facing backwards and bouncing gives the same sort of intensity as a PM6 in the forward position.

Two drivers were used in the Super Acousta, side by side, but emulating this with cheap drivers produces ultra directional sound, only relieved by two piezo tweeters facing 45 degrees sideways

Best wishes

David P

[Squeak]

So what are you saying? Is this a good idea or a bad? And should I go for horizontal or vertical placing?

[mondogenerator]

ive wondered something similar before but never tried it.

Are you thinking 2 drivers 'focussed' on the listener, 'toed in', or the opposite, and 'toed out' ? either way, i should imagine the angles would be tiny, and you could merely use foam sealing tape under the driver frame to seal, and vary angle using washers under the frame. shimming up on side by 1mm-6mm might be all you need to get a desirable effect. i would also suspect, the larger the angle the greater the spacing of the lobing will become. ie small angle = lots of interferance effect grouped closely, and bigger angle = wider spacing of lobes. which one in practice sounds better, i can only guess at! personally i would try for the smallest angle you can though, since bigger angles, i think, would mean that listening distance then becomes a factor in tonal stability, as with multiway systems but over a smaller bandwidth. since these are fullrangers, i would imagine this is something to avoid like the plague.

in the name of experimentation i would try the same box vertically and horizontally. This is mainly because, either way you are going to get lobing right in the 0 degrees 'sweetspot' as well as numerous others (measurable or not) in the 'typical' listening ranges of 45-0-45 degrees.

i have wondered about 3 FR drivers. this was mainly to allow the shaping of directivity through a designed bandwidth. one driver on each of 3 panels of a box. the 'side' drivers are -90 and 90 out of phase. 1st order bandpass on each of the driver in the side panels, and forward driver running true FR. the idea being to cause the side drivers to have a resultant 180 phase shift, and cancellation of some of the front firing drivers large angle radiation over a given BW, and variable by attenuation of these side drivers. although a variable phase and BW would be even better. sounds like i need to go active to experiment with THAT idea....

[BHTX]

Know anyone who has bose cubes with the dual vertical arrangement? (they're all over the place) Might be good to experiment with.
I'm visiting a family member right now who has been using this acoustimas junk for the last 15+ years. I was just toying with this exact idea and experimenting with directivity and lobing etc with the center speaker earlier before I came across this thread.

[Squeak]

Originally Posted by mondogenerator

Are you thinking 2 drivers 'focussed' on the listener, 'toed in', or the opposite, and 'toed out' ? either way, i should imagine the angles would be tiny, and you could merely use foam sealing tape under the driver frame to seal, and vary angle using washers under the frame. shimming up on side by 1mm-6mm might be all you need to get a desirable effect. i would also suspect, the larger the angle the greater the spacing of the lobing will become. ie small angle = lots of interferance effect grouped closely, and bigger angle = wider spacing of lobes. which one in practice sounds better, i can only guess at! personally i would try for the smallest angle you can though, since bigger angles, i think, would mean that listening distance then becomes a factor in tonal stability, as with multiway systems but over a smaller bandwidth. since these are fullrangers, i would imagine this is something to avoid like the plague.

Most larger retail speakers today use two or more drivers for mids and lows, one above the other. So I can't really imagine it being a great problem. My scheme would get rid of most of the HF lobing by pointing the beams in different directions and in the process getting better spread.
My understanding is that interference is mainly a problem at higher frequencies.

in the name of experimentation i would try the same box vertically and horizontally. This is mainly because, either way you are going to get lobing right in the 0 degrees 'sweetspot' as well as numerous others (measurable or not) in the 'typical' listening ranges of 45-0-45 degrees.

I really don't see any advantage of doing it horizontally. Only disadvantages.

i have wondered about 3 FR drivers. this was mainly to allow the shaping of directivity through a designed bandwidth. one driver on each of 3 panels of a box. the 'side' drivers are -90 and 90 out of phase. 1st order bandpass on each of the driver in the side panels, and forward driver running true FR. the idea being to cause the side drivers to have a resultant 180 phase shift, and cancellation of some of the front firing drivers large angle radiation over a given BW, and variable by attenuation of these side drivers. although a variable phase and BW would be even better. sounds like i need to go active to experiment with THAT idea....

That has already been done in a commercial speaker, with two speakers set close together, oriented horizontally. There was a special circuit that compared the two signals. Where on was different from the other, an out of phase but otherwise identical signal would be send to the other speaker. That way you could have good stereo imaging in one cabinet, by effectively cancelling one half of the drivers output.
I don't remember the make, but it was a white cabinet with black round metal mesh grilles.