What are the effects of a custom D'Appolito array that is both vertical and horizontal?
ie, two woofers horizontally and two woofers vertically both sharing the common center tweeter.
Something like:
*M*
MTM
*M*
(Please ignore the '*'s in the above diagram, spaces at the beginning of the line were ignored)
Just wanted to explore both benefits and disadvantages of such an arrangement.
ie, two woofers horizontally and two woofers vertically both sharing the common center tweeter.
Something like:
*M*
MTM
*M*
(Please ignore the '*'s in the above diagram, spaces at the beginning of the line were ignored)
Just wanted to explore both benefits and disadvantages of such an arrangement.
This 2-Dimensional Horbach patent paper goes in depth of the question:
LOUDSPEAKER ARRAY SYSTEM
Follgott has already set up a beautiful 2-dimensional Horbach-Keele Array.
LOUDSPEAKER ARRAY SYSTEM
Follgott has already set up a beautiful 2-dimensional Horbach-Keele Array.
A D'Appolito array uses a specific combination of spacing and crossover slope to narrow the vertical beamwidth of a two-way array. You can simulate it fairly quickly with xdir.
I think the work that Horbach and Keele did, as well as David Smith at Snell, was to extend the concept into a three way or even four way array. Google "horbach keele" or "snell expanding array" to learn more.
If you took a D'Appolito array, and added two more midranges horizontally, it would look a lot like the Donald North array. This one is way more obscure, but it there IS a patent on it, with all the info you need.
The North array does exactly what you would expect it to do: it makes the beamwidth symmetrical, both vertical and horizontal.
One other "neat" thing about a symmetrical D'Appolito array, is that you may be able to raise the output and the power handling a little. This is because you can bring the midranges in tighter.
For instance, a pair of four inch midbasses have about as much output as a 6" midbass, but the 4" midbasses can be packed tighter.
I've posted a bunch of articles on diyaudio about this stuff, so check my post history or use Google to dig them up.
Horbach-Keele arrays require some waveguide on the tweeter, as practically sized arrays only work up to say 3 kHz. Their crossover frequencies (for a given driver diameter) are lower than that of a regular MTM.
But we are skipping the important part: why do you want to use an MTM array? Do you need the narrow directivity?
Yes: Pseudo-coaxial with narrow directivity (and Horbach-Keele filters)
But we are skipping the important part: why do you want to use an MTM array? Do you need the narrow directivity?
Last edited:
TBTL makes a good point. Most of these modern MTMs use tighter spacing that the D'Appolito arrays from the 90s.
Even the Snell expanding array, from Smith, uses tighter spacing.
The Horbach Keele paper suggest a spacing of 55% a wavelength.
For instance:
Let's say we have a MTM with a xover of 2000Hz. 2000Hz is 6.75" long. 55% of that is 3.7125".
From that, you can see that a xover of 2000Hz is impossible.
So you have to keep dropping the xover point until it works.
For instance:
Let's say we have a MTM with a xover of 1350Hz. 1000Hz is 10" long. 55% of that is 5.5".
This spacing is "do-able"; you can achieve it using 3" midranges. Possible even with 4" midranges, depending on the geometry.
But you get the general idea; in order to get well behave response off-axis, the spacing has to be tight.
Even the Snell expanding array, from Smith, uses tighter spacing.
The Horbach Keele paper suggest a spacing of 55% a wavelength.
For instance:
Let's say we have a MTM with a xover of 2000Hz. 2000Hz is 6.75" long. 55% of that is 3.7125".
From that, you can see that a xover of 2000Hz is impossible.
So you have to keep dropping the xover point until it works.
For instance:
Let's say we have a MTM with a xover of 1350Hz. 1000Hz is 10" long. 55% of that is 5.5".
This spacing is "do-able"; you can achieve it using 3" midranges. Possible even with 4" midranges, depending on the geometry.
But you get the general idea; in order to get well behave response off-axis, the spacing has to be tight.
Yes, I'm looking to have much directed/focussed sound so that we get as less interaction with the room as possible.
But again I don't want the speakers to have a headphone effect.
I have just started learning about speakers and room interaction. Not an expert in any direction. Also trying to understand crossover issues in parallel.
Still not decided what design to zero in on.
My Goal is to experience the best imaging possible.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.