Integrated 2ch sound field creation speaker (3" Full-Range x 6 pieces)

I built a front 2-channel surround speaker.
The drivers used are SPL Limited's LSE77-20R (4.5 USD).

Box name is MTX-0806 (TQWT with Damp-Duct).
The front is L+R, the left side is L-R, and the right side is R-L.
The surround effect varies from source to source, but the surround effect is not as good as expected.
I am wondering how to improve the surround effect in the future.

MTX-0806-03.jpg


MTX-0806-05.jpg
 

Attachments

  • MTX-0806-02.jpg
    MTX-0806-02.jpg
    205.3 KB · Views: 66
  • MTX-0806-01.jpg
    MTX-0806-01.jpg
    170.2 KB · Views: 71
  • SPL  LSE77-20R.jpg
    SPL LSE77-20R.jpg
    221.3 KB · Views: 66
  • MTX-0806-08.jpg
    MTX-0806-08.jpg
    72.6 KB · Views: 58
  • MTX-0806-07.jpg
    MTX-0806-07.jpg
    47.2 KB · Views: 58
  • MTX-0806-06.jpg
    MTX-0806-06.jpg
    164.5 KB · Views: 64
  • MTX-0806-04.jpg
    MTX-0806-04.jpg
    51.1 KB · Views: 68
Hi tansand.

The 3 and 4 driver systems are an electrical speaker matrix.
MTX-0806 is a “physical speaker matrix” and can be used with Class D and BTL amplifiers.

In the case of mono playback, with an electrical matrix the difference signals (L-R, R-L) become 0, but with a physical matrix all difference signals are output.

I believe that the surround effect is better with an electrical speaker matrix.
 
  • Like
Reactions: tansand
It seems that reproducing the inverse phase component in the same box will depress the low frequencies.
It seems to be a good idea to separate the difference signals in the box.

And, caution is required as the impedance is low.

MTX-0806-09.jpg
 

Attachments

  • MTX-0806-10.jpg
    MTX-0806-10.jpg
    105.1 KB · Views: 30
  • MTX-0806-11.jpg
    MTX-0806-11.jpg
    104.9 KB · Views: 29
  • MTX-0806-12.jpg
    MTX-0806-12.jpg
    103.5 KB · Views: 32
  • MTX-0806-13.jpg
    MTX-0806-13.jpg
    104.1 KB · Views: 32
  • Like
Reactions: tansand
There's a good collection of information in this post.
 
  • Like
Reactions: nandappe
the left side is L-R, and the right side is R-L.
It is a stereo field.

So the left would be left only , and the right would be right only.



These speakers are essentially 90 degrees off axis, so treble roll off would be as shown like any speaker 90 degrees.
Sound being directional anyways, off axis only 450 Hz on down would actually be audible. Or start becoming omni at 450 Hz
and be fully omni at 250 Hz down. Humans can still slightly tell direction at 450 Hz from additional brain function, not with ears.

To have rear speakers they would be mounted in the .....rear.
And only useful for movies or soundtracks were producer has placed actual tracks or sounds in the rear intentionally.


Normal " rear" speakers would be 180 degrees. Indicating the direction the emulated sound would be coming from.

For a " wide" stereo image, you just use normal 2 channel stereo, with the speakers set wide or far apart.

The so called "image" sounding "forward" or " rearward" is in the original recording.
If you want the image to move forward or rearward from the original recording.
Just flip the right / left.
 
Last edited:
  • Like
Reactions: nandappe
There's a good collection of information in this post.

Hi Karlsonate.

The URL you refer to is an electrical matrix with three drivers, so it is fundamentally different from the physical matrix presented in this thread.

It is a stereo field.

So the left would be left only , and the right would be right only.

physical matrix.jpg
 
There's a good collection of information in this post.
Thanks, I had missed this 2023 thread. Not sure if @ASCTim has somewhat changed his assessment; there were multiple threads on related subjects (such as his "A Hafler inspired solution for the phantom center image problem"). All rather confusing to me! Good to see his (2023 above) chart that higher frequency lost stereo separation in "differential" L-R | L+R | R-L, which made perfect sense and matched what I heard.

This below, "wall-bounce dipole-differential with omni-center console", using four drivers, two amps and normal wiring, might be closer in spirit to @nandappe's?
 
Last edited:
I changed the physical matrix wiring to a normal 2-channel wiring(3 parallel).
I thought the stereo feel would be worse, but due to the high sound pressure on the sides, the stereo feel is greater than the box size.
Bass has also improved.

I will be demonstrating this speaker at a presentation with my audio friends on April 27.

View attachment 1439811


View attachment 1439812

View attachment 1439813

I think this makes sense. You are now getting pure 2L bounce off the left wall etc., rather than a phase-randomized admixture with -R.
Short conclusion is that differential (L-R, 2L-R etc.) in principle reduces stereo separation at high frequency producing enhanced mono; then tricks are needed to recover stereo effect.

I would be even more intrigued to see your 3-driver bipole firing orthogonally front/side/upward. Maybe you can test that by placing a small wing to bounce one of the side drivers upward toward the ceiling.
 
Last edited:
  • Like
Reactions: nandappe
Of course. The whole system is equivalent to 2 drivers (front) in the box, low-frequency wise.
Loudspeaker box should be optimally designed for a fixed number of drivers only. For how much drivers is this box designed?

Hi Sonce.

My favorite way to have fun is to make boxes as I think of them and attach various drivers to them.
I don't have the skills to design and build a box specifically for a specific driver.