Is it worth building?

[eddantes]

I have an idea for a 3 way that looks alright with horizontal dispersion, but the vertical looks bad.

How important is the vertical and is this bad enough to abandon the idea. (I've tried many variations and it remains just .... meh).

 

[YouDontHaveEnoughBass]

if you satisfy both of these you can get away with bad vertical dispersion:
1) you only listen sitting down with your head on tweeter axis
2) you have thick carpet + high ceilings

honestly though that vertical looks a bit too bad regardless

 

[mhardy6647]

Listen to a pair of Quad ESL-57. They're sublime, unless your ears* are too high or too low. The tweeter panels have, if memory serves, 15 degree vertical dispersion.
So, if you can live with a circumscribed listening area (not necessarily head in a vise), you could, maybe, live with it.
If that's not the goal, though, probably worth investigating something else.

I would also say that a model is just that. If the resources and bandwidth exist to prototype, there's value in knowing that the sims (EDIT: I am assuming the "data" are sims?**) are speaking truth.
Coming from a life science perspective, the models that we can cherish and put on pedestals can be... non-comprehensive. How's that for a diplomatic way of putting it?

Good luck with your odyssey quest!
________
* EARS, not "unless you are too high..."
** If they're real data, then, yeah, I'd think twice before cutting wood or MDF, or casting concrete, or firing up the 3D printer... or whatever...

 

[eddantes]

Hmm... I have another MTM idea that looks like this

This looks better, but, it just doesn't go low enough.

 

[eddantes]

... the models that we can cherish and put on pedestals can be... non-comprehensive. How's that for a diplomatic way of putting it.

Ain't this the truth... Here I am wondering this question - when in reality of the 5 speakers I built only two were reasonably close to the model. The prototype is the thing.

 

[mhardy6647]

There is some value to F'ing Around and Finding Out... with certain exceptions, of course.

 

[Dillion]

I’d probably lean toward prototyping it before abandoning the idea. The vertical response looks concerning, but if the crossover and driver spacing are otherwise working well, it seems worth finding out how much of that survives in the actual speaker. Worst case, you learn something useful for the next iteration.

 

[Roland68]

I have an idea for a 3 way that looks alright with horizontal dispersion, but the vertical looks bad.

How important is the vertical and is this bad enough to abandon the idea. (I've tried many variations and it remains just .... meh).

View attachment 549917

View attachment 549918View attachment 549920

May I ask what kind of MTM configuration this is?
Have you tried angling the two mid-range drivers slightly towards the tweeter (so that their axes intersect only behind the ear)?
You could also try forming a sort of waveguide around the entire MTM block (relatively narrow and not too deep).

 

[eddantes]

@Roland68 - I can only choose one design to build - because all designs use the same drivers. I have considered a woofer supported MTM as well.

The MTM:

And the 3 way

4 Driver idea

Things never really get better. The MTM looks ok enough to attempt a prototype. Horizontal looks good for all three though:

 

[Roland68]

@Roland68 - I can only choose one design to build - because all designs use the same drivers. I have considered a woofer supported MTM as well.

The MTM:

View attachment 550162

And the 3 way

View attachment 550163

4 Driver idea

View attachment 550164

Things never really get better. The MTM looks ok enough to attempt a prototype. Horizontal looks good for all three though:

View attachment 550167

View attachment 550166

View attachment 550165

Are these purely simulations rather than actual measurements?
Is the spacing between the mid-range drivers—and between them and the tweeter—being taken into account?

Personally, I’m not really a fan of the MTM configuration—especially for a center channel—but that’s a whole other topic.
That said, my own main floor-standing speakers actually use an MTM design, with two woofers positioned below them.
For an MTM design, I’d recommend using mid-range drivers no larger than 13 cm, paired with two 19–20 cm woofers. This allows for a relatively narrow baffle, meaning you can apply a more pronounced radius to the edges next to the mid-range and tweeter drivers if you double up the front baffle thickness in that area.
You can also trim the front panel of the tweeter/make a round cutout so that you can bring the midrange drivers a little closer together; that can make a big difference.

 

[DanielT]

It is a challenge with MTM. Here is what a technical consultant in acoustics and audio engineering says about the principle:

With MTM configurations, in addition to narrower radiation lobes around the crossover frequency, significant irregularities in vertical dispersion arise when the M-to-M distance is comparable to the wavelength. This is the main reason why such configurations are uncommon today.
.....

Naturally, MTM configurations do not perform well at all regarding uniform vertical dispersion, but neither do the MT configurations found on the market meet all the criteria—except, perhaps, in extreme, exceptional cases.
.....

Once again, the interference issues have practically nothing to do with the filter's steepness, as they arise well below the crossover frequency—even if the center-to-center distance between the drivers is reduced to almost zero. Even a center-to-center spacing of 15 cm is too large, necessitating the use of very small drivers, which in turn leads to other problems. There are good reasons why the MTM configuration is relatively uncommon in commercial speakers.
........
Regarding the issues associated with MTM or BMTMB configurations, problems arise well below the crossover frequency—specifically when the center-to-center spacing between the dual drivers corresponds to approximately half a wavelength (or even less, strictly speaking). In a typical MTM system with 30 cm spacing, a massive dip occurs at 45 degrees off-axis (roughly corresponding to the ceiling reflection) as low as 500 Hz. Significant problems also arise at shallower angles, such as for a standing listener.

https://www.faktiskt.io/phpBB3/viewtopic.php?f=3&t=72656&p=2208978&hilit=MTM#p2208978

It should be added that, in the thread I am quoting, he set—at the request of others—fairly high SOTA level/standard.

 

[eddantes]

The 165mm drivers are ~150mm from 0 so 300mm apart. So that's about 1.1k wavelength. Not sure what I should make of it. I just attempted to place them as close together as the drivers would allow.

 

[DanielT]

Why not try out the MTM design you’re considering, but—if you feel like it—also try converting it into a 2.5-way or 3-way speaker? See which you prefer.

Of course, there will be some extra costs for the additional crossover components, provided you use a passive crossover.