Midrange drivers in a circle?

I have never seen this kind of speaker that would have four or five midranges very close to each other and a tweeter in the center.

What would be the advantages and the disadvantages when compared to MTM for example?

hsapril00A.gif

circles-in-circle-problem.png


It would be also possible to tilt the drivers pointing more inward or outward, how would that change the situation?
 
I would be inclined to eliminate the version with five in a circle, it could have some odd lobing issues.
As for the other one, you could tip it up so the mids are in a horizontal/vertical arrangement and treat it like D'Appolito set up. It's an interesting idea.
As to tilting the drivers in or out, that would depend on the dispersion characteristics of the drivers, and the wavelengths they reproduce to determine what, if any effect it would have.

Mike
 
Last edited:
Unless you crossed over to a tweeter at an unusually low frequency you would see significant beaming.

Without resorting to math, you can view this as one driver composed of the four (or five).

Assuming 4 to 6 inch drivers it would be broadly similar to driving an 8 or 12 inch driver at those frequencies.

A d'appolito gets away with things, indeed utilises this, by setting the drivers vertically. The horizontal radiation is as per the individual drivers, but the vertical is squashed in due to the larger vertical radiation size.
 
I have never seen this kind of speaker that would have four or five midranges very close to each other and a tweeter in the center.

What would be the advantages and the disadvantages when compared to MTM for example?

hsapril00A.gif

circles-in-circle-problem.png


It would be also possible to tilt the drivers pointing more inward or outward, how would that change the situation?

I think you'll dig this : Optimal MTM Geometry
 
Machismo, there are many speakers with such arrangement. You just did not search hard enough. Some even very successful commercially. Legacy audio...whisper.

Not so much the Whisper since the the arrangement is not circular:

806legacy1.jpg


But the Legacy Audio Helix looks just what I was looking for. And it turned out that they have tilted the midranges just like I thought might be necessary. Also the pack is tight.

helix.jpg


Kharma Veyron EV-1 has also something like this, but more on the tweeter side.

Kharma-Enigma-Veyron-Loudspeaker-System-2.jpg
 
Unless you crossed over to a tweeter at an unusually low frequency you would see significant beaming.

This is the whole idea. Pattern control. Horn like beaming without a horn. Seeing the tekton design with all the tweeters I shook my head and thought it was a terrible idea. Then at rmaf I heard a pair and was shocked how horn like the sound was. When the guy in the room said that the middle tweeter was running higher frequencies and the surrounding ones were operating lower, the light bulb went on as to what they were up to. They sound good, just an expensive way to get pattern control.
 
I played around with 2 3.3"drivers and a tiny dayton mini6 planar tweeter, 5k x-over in this configuration


warpoperat-En.gif


compared to the 2 drivers stacked and the tweeter in front in the middle, the sound had changed to the sound you get when bigger drivers play higher frequencies; It sounded bigger but not better, less 'precise'.

The other thing I did not like was the higher power response to the sides (very equal as in front) as it made me have to sit exactly in the middle of them; toe-in in front of the listener did not work for the mids with this setup.

Here I got the idea that drivers should not be to wide for the frequencies they play. They start to beam as well.

I started to wonder if bigger drivers get more directional due to their own comb filtering. Which with smaller drivers used in the horizontal plane, it is widely accepted as being unwanted.

If so, we should not us drivers wider then 1 (or maybe 2?) times the wavelength they emit/play.
I.e 4100Hz ~3.3" - 1700Hz~ 8" - 1150Hz~12" (actual radiating surface is a bit less).

Wavelength




To me a few smaller drivers in the vertical plane is more favorable then using 1 bigger driver so far. You have so many advantages with i.e detail resolution, directivity, power compression, no x-over from 150Hz-5KHz (I guess very important to the great SQ), baffle compensation is easy as it only affects the mid drivers. A wooden baffle as narrow as possible and 1 piece of pvc pipe as sides and back seem to do very well.
 
Last edited:
Follow the link in PB's response - Optimal MTM Geometry. I believe it will help to answer your query.

I see that folks mentioned Legacy speakers. I have heard the Helix several times and was really impressed. They have many drivers, the bass drivers are isobaric and powered. There are 4 mid-ranges (which I thought was not a good idea) but I really liked them. But my other take away is that they mainly for large rooms.
 
Last edited:
  • Like
Reactions: 1 users
So there was one speaker after all... Thanks for the link!

However, it does not seem that it was sold very much?

Donald North, owner of DNA Audio and owner of the patent on this, is on the forum. Drop him a line!

What do you think about the extra gaps, do you think they are only because of better looks or would making a tighter pack like this ruin something?

An externally hosted image should be here but it was not working when we last tested it.

Arguably, the main advantage of this layout, versus a conventional coax, is that you can manipulate the center-to-center spacing to change the beamwidth. For instance, if you used very very small midranges and packed them very tight, you could get something close to 180 degrees of beamwdith. As you widen the center-to-center spacing, your pattern is going to widen.

As I see it, there are five good sources of info on this:

1) the Donald North patent

2) the Horbach-Keele paper over at xlrtechs.com

3) The Horbach-Keele thread here

4) "Crossovers, a step Further" by LeCleach which everyone should print out and put up on their bookshelf because its just 100% great

5) my thread has bits and pieces of the three above

Whatever you do, don't even consider this project unless you have some decent software to simulate this. To get good results you have to manipulate the xover points, the slopes, the phase, the spacing... it's a lot of variables.

-1x-1.jpg

If you really have a lot of time to kill, you can manipulate the baffle shape too. The Beolab 90 is curved for a reason...
 
Last edited:
Unless you crossed over to a tweeter at an unusually low frequency you would see significant beaming.

Without resorting to math, you can view this as one driver composed of the four (or five).

Assuming 4 to 6 inch drivers it would be broadly similar to driving an 8 or 12 inch driver at those frequencies.

A d'appolito gets away with things, indeed utilises this, by setting the drivers vertically. The horizontal radiation is as per the individual drivers, but the vertical is squashed in due to the larger vertical radiation size.

No.