As a follow-up to my previous post, here is a further elaboration of the sphere concept with four woofers and four tweeters.
The drivers are evenly distributed around the sphere (left image).
On the right is the placement of the drivers in the Boxim simulation. Not perfectly equal, but I hope good enough to recognize certain trends.
Simulated are 3 inch woofers and 3/4 inch tweeters on a 20 cm sphere. Drivers were ideal apart from their cone/dome diameters. Drivers were crossed at 3 kHz.
Two situations were simulated:
In the first simulation, none of the speakers was delayed.
In the second simulation, the tweeters and three of the woofers are delayed such that all the drivers are virtually in the same plane. This plane is perpendicular to the listening axis.
It appears that the SPL on the listening axis can be quite flat. However, the directivity plot shows that the range of flat SPL is quite small for high frequencies. I suspect that this means that only one person can listen to stereo properly. Also, the speakers must be accurately aimed on the sweet spot.
I have little experience in interpreting directivity plots but this result does not seem optimal from the perspective of accurate stereo imaging. Therefore, comment on these results are very welcome.
The drivers are evenly distributed around the sphere (left image).
On the right is the placement of the drivers in the Boxim simulation. Not perfectly equal, but I hope good enough to recognize certain trends.
Simulated are 3 inch woofers and 3/4 inch tweeters on a 20 cm sphere. Drivers were ideal apart from their cone/dome diameters. Drivers were crossed at 3 kHz.
Two situations were simulated:
In the first simulation, none of the speakers was delayed.
In the second simulation, the tweeters and three of the woofers are delayed such that all the drivers are virtually in the same plane. This plane is perpendicular to the listening axis.
It appears that the SPL on the listening axis can be quite flat. However, the directivity plot shows that the range of flat SPL is quite small for high frequencies. I suspect that this means that only one person can listen to stereo properly. Also, the speakers must be accurately aimed on the sweet spot.
I have little experience in interpreting directivity plots but this result does not seem optimal from the perspective of accurate stereo imaging. Therefore, comment on these results are very welcome.
I have no plans to continue this thread for now.
Reasons are the questionable value of the simulation of an octahedron on a cube, the not very great stereo performance and the great complexity of the last system presented.
Still, I keep looking for a simple system with both good omni directional sound reproduction and also good stereo performance.
I started a thread on the multi way forum proposing a system where two midwoofers and 2 tweeters are housed in a cube shaped enclosure.
If you’re interested, visit https://www.diyaudio.com/community/threads/2-way-omni-cube-speaker.417265/#post-7792510
Reasons are the questionable value of the simulation of an octahedron on a cube, the not very great stereo performance and the great complexity of the last system presented.
Still, I keep looking for a simple system with both good omni directional sound reproduction and also good stereo performance.
I started a thread on the multi way forum proposing a system where two midwoofers and 2 tweeters are housed in a cube shaped enclosure.
If you’re interested, visit https://www.diyaudio.com/community/threads/2-way-omni-cube-speaker.417265/#post-7792510