Acoustic Horn Design – The Easy Way (Ath4)

[Mindsource]

Not as impressive as I had thought they were... And that was done just to get closer CTC?

 

[gedlee]

Hence my disapproval.

 

[mabat]

Yeah, it's the vertical I would assume the worst in this case. This never works as intended (such high aspect ratio). Thank you!

 

[gedlee]

In a square waveguide the vertical will improve to match the horizontal, and the diagonal will not improve much, if at all.

 

[mabat]

Actually, the diagonal can be better than both, H and V. Seen that many times in the simulations. Don't ask me how is that possible.

 

[Vathek]

Isn't that about what one would expect an SEOS to perform like given the extreme asymmetry though?

 

[Mindsource]

Yes, lol. I may have had some willful ignorance towards their performance. When they came out there weren't many WG's that a consumer could buy, except maybe the QSCs. And all you would see were the horizontals, which were fine. But in hindsight, they were an overall lacking design.

 

[Vathek]

The perform just fine if their design decisions are what you want, I just think the asymmetry is too extreme. And there still aren't any available alternatives really with the QSC out of production. Which is a shame as the 90 x 60 pattern seems more reasonable to me.

 

[mabat]

I have yet to see a waveguide such as 90 x 60 that is really 60 deg in any usable sense. Too often this means just a rising DI.

 

[gedlee]

Actually, the diagonal can be better than both, H and V. Seen that many times in the simulations. Don't ask me how is that possible.

Show me.

 

[mabat]

Show me.

Perhaps I could find some more illustrative example but this one I have at hand (I haven't done a rectangular WG in a long time): #2494
- Doesn't it have the best behaviour along the diagonal?

 

[mabat]

I could scale it down to 1" (it is 1.4" originally) but that would still be 298 x 257 x 96 mm. Is there anyone able to print that for testing?
Or what about 0.75" or 0.5"?

Maybe I would also increase the 'k' parameter of OS-SE. Now I usually make it slightly > 1 and didn't have this paramater at that time.

 

[gedlee]

Perhaps I could find some more illustrative example but this one I have at hand (I haven't done a rectangular WG in a long time): #2494
- Doesn't it have the best behaviour along the diagonal?

No, not at all, its DI is higher than any other direction, exactly as I would expect. Not what I want.

 

[mabat]

Why? How much is it higher and what's wrong with that?

 

[AllenB]

I have yet to see a waveguide such as 90 x 60 that is really 60 deg in any usable sense.

Do you mean one that actually reduces ceiling reflection levels throughout its whole band?

 

[gedlee]

Why? How much is it higher and what's wrong with that?

This is the same old argument - I want a flat DI, that device isn't.

 

[mabat]

Do you mean one that actually reduces ceiling reflection levels throughout its whole band?

Yes, one with a reasonably flat DI through the intended band.

 

[mabat]

This is the same old argument - I want a flat DI, that device isn't.

I'm pretty sure it is.

 

[gedlee]

It can't be. All directions will have about the same DI at LF - they have to. So if the DI in one direction is higher than the others at HF then the total DI has to be rising.

 

[JSS]

I could scale it down to 1" (it is 1.4" originally) but that would still be 298 x 257 x 96 mm. Is there anyone able to print that for testing?

I have a 1.4" bms 4594he if the throat match? I could print it maybe in quarter sections.