Acoustic Horn Design – The Easy Way (Ath4)

The same for the DFM-2535 (one-piece version added to the package T520-25-STD-3).
This is how it's printed (with some supports, of course):

1740209157031.png
1740210055999.png
 
Last edited:
Here is the raw measurement I got from the new horn: The dip at 12k is concerning - crossover is 4th order at 650hz (and 85hz to a subwoofer) I will measure again after sanding and painting. Just from initial listening I feel like that xover is a bit low for the CD (SBA 65CDNT)
1740275807405.png
 
Last edited:
I retuned it for 845hz 2nd order crossover today - much better imo. 3 petals to go on the 2nd horn!

I will bet you guys disapprove but all i do it the basic crossover (845hz 12db/o) and let dirac fix up the rest (i apply some curves in there) - it does a great job tbh.
 
These are simply not measurements which you can base any crossover design on. You need to take (semi)anechoic data, complex frequency responses (i.e. magnitude and phase) of all the drivers separately - that's an absolute must. Otherwise it's just a cut&try, unlikely to lead to anything really good. You're still basically blind here.

Dirac can "fix up" only a small part of it.
 
Last edited:
I didnt think it would work either but it actually sounds really good.

this is from REW with 1/6th smoothing after dirac (in room) . Yeah it is very far from flat but it sounds great tbh. I will put some more effort into it once the horns are smooth and painted and there is a pair of them.

I also time aligned each driver in the hypex amp and you can't tell sound is coming from the speakers it's just floating in space and sounds very real.

Green is your horn design mabat and purple is the PAP wood horn

1740297846731.png
 
Last edited:
  • Like
Reactions: 454Casull
Having printet 5 A520 horns now, each with different kinds of infill and number of outer walls (and different slicing - i really do prefer dowels to the original slice) I have come to the conclusion that the horn needs some kind of damping of bell modes. Need to measure the resonances but just tapping the horn it sounds too lively. This may just be me and my audiophile nevrosia. But even as the horn with 5 outer layers and 15% gyroid infill seems to have some kind of "deadish" quality, it still does not feel inert nor "dead" enough. Haha, hows that for beeing subjective.

Having prepared a horn for filling with epoxy but really not looking forward to the process I realized that having focused only on fillling the petals and center piece made me blind for the obvious solution. To make pockets for damping material. The standard way of dampening plinths for record players.

My next A520 will have petals looking like this. Its a A520 sliced in 6. Will fill the cavities with epoxy granite or maybe just plain silicon chaulk with steel shots. Printing for maximum stiffnes and with this way of damping I expect the horn to feel inert and "dead". Using the word "feel" twice now I think I may have gotten expelled from that other forum..

Petals with pockets.jpg
 
I didnt think it would work either but it actually sounds really good.



View attachment 1426215
actually, it is quite good. With lowering bottom part (below 40) and couple of eq touch in upper range it will come to very flat freq resp. check the phase and play with delays as/if needed and you are in audiorvana. do not wory about small roughnesses, nobody can hear it, specially with music in stereo.
Should be very easy with Hypex
 
Having printet 5 A520 horns now, each with different kinds of infill and number of outer walls (and different slicing - i really do prefer dowels to the original slice) I have come to the conclusion that the horn needs some kind of damping of bell modes. Need to measure the resonances but just tapping the horn it sounds too lively. This may just be me and my audiophile nevrosia. But even as the horn with 5 outer layers and 15% gyroid infill seems to have some kind of "deadish" quality, it still does not feel inert nor "dead" enough. Haha, hows that for beeing subjective.

Having prepared a horn for filling with epoxy but really not looking forward to the process I realized that having focused only on fillling the petals and center piece made me blind for the obvious solution. To make pockets for damping material. The standard way of dampening plinths for record players.
Yeah interesting idea - You could do a lot of damping with silicon I would think - assuming it bonds well to the plastic.

For what it is worth I can't hear any resonances. (with music) But I do agree the cones with 15% infill don't seem to be as solid as one might like.

I just went tapping on it vs tapping on my wood horn and the plastic horn is deader than the wood one. Surprising to me but it's probably the glue between the petals.

Cubic infill might be deader as a choice of infill since it is all sealed pockets of air? Not sure how that would play out. I believe there was a youtube person testing this and found very little differences in infill and thickness.
 
Last edited:
What if the below dip is there because the horn and the woofer(s) are cancelling each other, i.e. they are not well in phase?

1740328925525.png


What do you think can the Dirac do? It will only increase the voltage in that region. Instead of solving the issue it will only drive the transducers harder.
Then you come and say that the driver needs to be crossed higher... Well, when treated like this it probably does 🙂
 
Last edited:
Yes - one need to first linearise (EQ) the drivers well in their (to be) stop bands before applying the crossover - at least 1 octave - so when bass is flat up to 2x XO and CD is flat down to XO/2, apply the XO - if no severe resonances etc is present the minimum phase situation will see to that the acoustic phase will cooperate and XO will work as intended.

//