Acoustic Horn Design – The Easy Way (Ath4)

[mabat]

FIFA might sue you, though...

Soccer ball is not made of hexagons (only). That's not even possible

 

[mabat]

BTW, for the Peerless DFM-2535 (see the group buy), I took the A460G2 and let it optimize again for this particular driver, this time releasing the constraints for the common Gen2 body. Here's the result, it's a tiny bit less narrowing towards high frequencies. Not sure yet what to make from it, I'll probably just upload the profile template/DXF to Cults.

 

[Psytron]

It may not look so but it's all printable without any supports. Quite a care was put into this.

Yep, I was definitely missing that. Well done.

1. love the mortis and tenon. The loose tenon is a good idea if they need a little fitting to make a snug fit its easy to do.
2. I would like to know why you have the "crown" part thin out before terminating?

 

[Psytron]

You may find these posts of interest:

This is what I love about communities on the internet. Super cool by the way!

Have you done any trials lately? do you think its worth it?

 

[mabat]

2. I would like to know why you have the "crown" part thin out before terminating?

It's just what's possible, but I'm going to rework this detail and add a thread that will tighten up the parts when glued together.

 

[Psytron]

I usually find I can get the smallest seam when the cut is normal to the face I am cutting.

 

[Speedysteve7]

I've got a few other variations of this that I've printed and tested, but so far this one looks to be the most promising, so I'll start on a physical prototype next week.

This looks like the next logical step forward in the evolution of the pyramid MEH.

It looks fantastic.

Please let us know how it goes and if 3d model etc will be available?

 

[Speedysteve7]

Also, is it just me or does this configuration scream to be used on a coaxial driver?

I've been coaxial curious for a while but the SQ I get from 4 MF drivers / the usual compression driver atop, plus being able to try different compression drivers to taste and the price of good coaxials (I've never heard any either other than a BMS4592 neo) has dampened that curiosity.

This is why this format really piqued my interest.

 

[mabat]

I usually find I can get the smallest seam when the cut is normal to the face I am cutting.

OK, I think I see now what "crown" did you mean. Well that's a consequence of cutting in hexagonal grid. I don't see it as a problem.
The smallest seams are, IMO, for cuts in planes parallel to the bed, but that's not always possible.

 

[mabat]

BTW, for the Peerless DFM-2535 (see the group buy), I took the A460G2 and let it optimize again [...]

This is the sim (normalized@5deg). It should be pretty much the same in real.

 

[tmuikku]

Hi, nice lookin. I'm curious which angle is the meandering trace highlighted here? looks like it's something like 170 deg? Unusually high top octave for high off-axis angle.

 

[mabat]

That's 180 deg, there's always this narrow, low-level "beam" right behind.

 

[tmuikku]

Asymmetric waveguide would drop that mixing path lengths, if it's important. On the other hand strong constructive interference there indicates good profile, sound diffracting nicely around not losing much intensity to diffraction back wave.

 

[mabat]

I consider this pretty much irrelevant. In terms of radiated power, this is just nil.
It's included for completeness, mainly to calculate the DI in the proper way, in principle, but otherwise it just doesn't mean anything, IMO.

 

[tmuikku]

For sure, I found it as an interesting detail in the sim. Although, most interesting detail is the top octave overall, if it measures like sims it doesn't get better than that.

 

[mabat]

Above ~15 kHz it depends on the driver in the end, on the coherency of the produced wavefront. There will be only a few that work so well that high. But then, there's no reason why it shouldn't measure as predicted. Up to 15 kHz, it virtually always matches very well (for a 1" throat).

 

[mabat]

This is a simulation including the exit conical section of the driver (DFM-2535), in case it was driven with a) flat and b) spherical wavefront:

a)

b)

The DFM-2535 is usable up to 17 - 18 kHz, this should be good enough

 

[mabat]

The latest A460 - throat impedances for the first two axisymmetric HOMs of the source.
They basically won't radiate at least up to ~12 kHz. With the DFM-2535, it seems there aren't any anyway...

 

[Olombo]

Okok....also took part of the group buy now 🤣

So if you publish the latest iteration for zhe DFM I would be interested.

 

[TNT]

So we know how to identify HOMs in the charts? I didn't think we understood that as HOM is not your every day regular resonance... as I understand it..!?

//