Acoustic Horn Design – The Easy Way (Ath4)

[wesayso]

I'd say there's more to test?

Looking at this picture:

Makes me wonder, what happens if...

- we make sure each channel has the same expansion rate
- build up expansion, the further the channel is from the center (thus making the middle one fall in between the center and outer channel)
- etc...

It will look strange, in 2D, but the important part seems to be the gradually growing expansion.
I'd make the studs that keep the channels fixed drop shaped . Round part toward the throat.

 

[mabat]

That makes no difference, I tried that already (didn't show). It seems it's only the total area that matters.
The widths/positions of the channels have effect on HF polar respose but not on the impedance, if that's what you mean.

BTW, it's definitely possible to "shape" the impedance by changing the shape of the waveguide - there's the room for optimization.
This is for a 1" driver -

 

[wesayso]

Good to know! It almost makes me wonder if there is a Synergy concept hidden in here. Adding the mid-woofers to the outer channels. But with this kind of behavior even for larger drivers... we might not need it.

 

[pelanj]

What are the dimensions of the object? This looks like a job for a resin printer - maybe this is finally a reason to buy one.

 

[mabat]

Well I already have a resin printer. Maybe it's time to try it... have been hesitating until now. I already bought a water washable resin. But that's mainly for a final production. I can prototype on FDM.

 

[mabat]

I tried to drive the same WG by an ideal spherical cap corresponding to the exit wavefront of the plug. Here's the result -

 

[pelanj]

I do not have one and I have been considering to buy one for a long time...but do not know which model would be the best for me.

 

[David McBean]

It is interesting that the attached two profiles look much the same, even though they were generated using quite different methods and theoretical assumptions.

 

[mabat]

I used just a random shape for now, basically a conical horn rolled back very early on.
The whole point is that you have to drive this profile with a spherical wave, otherwise it would be utterly useless (attached results for a plane wave).

At this point it's still not optimized in any way.

 

[kipman725]

This is all very exciting, what driver do you think would be a good candidate for this new phase plug? a 4" Be dome? (to avoid modal behavior of the diaphragm)

 

[aragorus]

You did it again Mabat!
Looks great!

It's the kind of loading one would expect from an exponential horn.

 

[aragorus]

So who is able to make it?

- I already tried a half on a regular FDM printer (0.25 mm nozzle) and at least for prototyping it's pretty good. The top is a single perimeter, so about 0.25 mm thick (would need some hardening though, maybe an epoxy coat, or just make the wall thicker). There are still also some minor slicer issues.

I'd love to give a 2" version with 4 vanes a go, if you'd provide a dxf or slices/points

 

[Patrick Bateman]

I do not have one and I have been considering to buy one for a long time...but do not know which model would be the best for me.

Creality Ender 3 is $250 USD and works really well.

I started out with their nicer model, and actually prefer their cheap one.

 

[sheeple]

I tried to drive the same WG by an ideal spherical cap corresponding to the exit wavefront of the plug. Here's the result -

What is it with the wrinkles in impedance plot, this is indicating resonances? Simulated SPL naturally shows the same variance. Can this be further reduced? What is the sonic implication?

 

[fluid]

Creality Ender 3 is $250 USD and works really well.

I started out with their nicer model, and actually prefer their cheap one.

That is not a Resin printer which is what Pelanj was referring to

Just noticed a good price on a Creality Resin Printer (no idea if they are good or bad) and they are available for Czech's

Creality LD-002R 3D Printer | LD-002R 3D LCD Resin Printer

 

[pelanj]

These are definitely available. But there are many more. The print volume is the key. Is the "cheap standard" volume enough or do I need more for this? I have 3 FDM printers. Prusa Mini is the best of them.

 

[mabat]

I'd love to give a 2" version with 4 vanes a go, if you'd provide a dxf or slices/points

DXF attached.

Public Fusion link: Fusion

Let's see.

 

[ScottG]

I do not have one and I have been considering to buy one for a long time...but do not know which model would be the best for me.

The largest volume you can afford..

About 6 months ago the volume "bumped" from the phone-screen size printers to the aprox. 8" printers. I purchased one of those (AnyCubic Mono X) but now wish that I had waited for the next "bump" in volume (at a fairly reasonable price).

Here is the latest "bump" in *volume vs price:

ELEGOO Jupiter: 12.8" 6K Mono MSLA 3D Printer by elegoo official — Kickstarter

Be aware that other manufacturers like AnyCubic often produce a similar product near or soon after when a product like the Jupiter actually starts shipping, so it's quite likely that several months from now AnyCubic will have an initial sale on a similar size printer; this is because the screens they are using (12.8" in this instance) become available for bulk purchase. Caution: with the global chip shortage - it's anyone's guess if the product will be actually ready to ship to most customers at any given point until that problem is resolved.

*note: you can purchase even larger volume resin printers, but most are around the 2k price-point with a few that are more expensive.


Of course you can slice a larger model and then resin "glue" (or super-glue) it together after printing the multiple pieces, but it's a messy hassle.

 

[mabat]

What is it with the wrinkles in impedance plot, this is indicating resonances? Simulated SPL naturally shows the same variance. Can this be further reduced? What is the sonic implication?

I just don't know (yet). I think it will take a while now to trace all the little remaining imperfections. It's already pretty complex structure.

Any ideas?

 

[bappe]

What is it with the wrinkles in impedance plot, this is indicating resonances? Simulated SPL naturally shows the same variance. Can this be further reduced? What is the sonic implication?

Can it be an impedance discontinuity where the "phase-throat" hands over to the WG?

//Anders