3D printed waveguides

[augerpro]

Here is the profile I'm working on. Makes me wonder how so many commercial designs get away with such shallow throats and no phase shield.

[augerpro]

Alternately you can flip the ellipse so the wall at the throat is straighter and tighter curve is happening at the mouth:

[TN Allen]

Brandon,

Would you include cross sections?

I may have a few elliptical guides you can try, depending upon the throat diameter you want.

It may be worthwhile, though a bit of a challenge, to produce a throat geometry that will allow for different diameters to accomodate different tweeters. This is challenging and limiting though because there must be a throat transition area that blends the elliptical contour into a circular contour that can be "sliced" off for different diameters. This significantly changes the curves defining the guide surface, making them more complex, but also allowing the use of various tweeters.

There may be intetesting accoustical effects as well, which would need careful evaluation.

This is much more easily done with round guides, where cutting off, or adding to the thickness to change the throat diameter maintains a circular throat. In simple elliptical guides, reducing or extending the thickness alters the throat producing an elliptical throat opening, rather than circular.

The above is just an idea to consider, perhaps impractical, but if workable, could make the elliptical guides more versatile.

[Pete Schumacher]

I'm not a fan of having a mouth opening narrower than 90 degrees. With a very steep wall at the throat there is more chance for the formation of reflections side to side as the wave front travels out the guide. Keeping the exit angle at 90 degree (45 degree walls) or larger, you eliminate the chance that a reflection off the wall will be able to head back across the guide and reflect off the other side.

[augerpro]

I apologize, I haven't been using the standard language for horns, so I've been calling out the wrong angles, which might have led to some confusion. For the two tangent mouth waveguides, throat entrance angle (wall to wall) is about 70 degrees. For the non-tangent wg's the throat angle is closer to 90 degrees. For the last two drawings I posted I don't have exact measurements, but just eyeballing it it is probably around 60 degrees.

Pete> I agree I don't like walls too steep, but it does appear that the deeper wg's others have posted worked well, and *probably* have a steeper throat than mine. The mouths on the other hand, are all pretty wide opening, greater than 90 just eyeballing it.

Tim> I mispoke, those are not elliptical waveguides, they are circular, but an ellipse is used to define the wall flare or curvature. The others were constant radius wall curves, i.e. a circle.

Does anyone see any issue with the last two designs I posted using an ellipse to define the wall flare? I'll probably have both printed.

[TN Allen]

Brandon,

I realized this morning that the ellipse shown in each image is the cross section, rather than a top view. That is an interesting alternative to using arcs. Why not try these? You won't know how they effect the sound otherwise. Ultimately how they sound projecting music is more important than the shape of the test results.

I suspect that guide shape can vary quite a bit, and still have a pleasant effect projecting sound. Some of the contorted double guides I've milled work quite well, though their test results would probably indicate they should be dismal. Of course that "quite well" is very subjective.

Hopefully this link to some interesting horns(not mine) that illustrates the extreme shapes people try, works:

https://goo.gl/images/qnoHU2

[augerpro]

Just did three more designs, to test throat angle. Basically they are all conical so I can set that angle, then with huge fillet that smoothly transitions from the conical throat to the baffle. I designed for a 80, 90, and 100 degree throat entry angle.

[TN Allen]

Brandon,

I think in an earlier WG thread there may have been some discussion about spiral cross sections. A spiral, possibly also Fibonacci curves might prove useful to produce a suitable throat and mouth configuration.

I've sometimes wondered how a cross section combining convex and concave curves might affect the response.

[augerpro]

Well I originally used a golden ellipse, so that should be the same as what you describe. But what i realized was that, once you define the tangency, depth, mouth diameter, and throat diameter, there is only one curve that is allowed, so you can't define it. You have to give up one of those things if you want the degree of freedom back, in this case it would probably be tangency, although you could give up depth if was within reason too.

[augerpro]

I've sent 4 designs to the printer. One has the surround overhang, and really I probably should have just done this one since if it works well, would be a feature that is carried over to the other designs. The other two are posted just above using an ellipse to define the wall flare, using two different orientations on the ellipse to change whether the sharper curvature is happening at the mouth or the throat. The fourth wg is an actual elliptical, horizontally it is the same as the .75" tangent mouth waveguide I measured.

[Pete Schumacher]

Just for quick reference, here are the on axis response of the guides I used on the RS28A. The first is a 2.0" deep guide about 7" diameter and fairly steep throat walls, not sure exactly of the angle.

The second is with the 1" deep version with 8" diameter mouth. Notice the obvious differences in the top octave. It was audible too, even to these older ears.

[augerpro]

Were these conical at the throat, or curved? If curved what was the radius?

[TN Allen]

Pete,

It's great that you listen as well as test. I only listen, and have no standing regarding testing. Nonetheless, I think WGs offer some wonderful benefits for "older ears", mine included. Hopefully, as Brandon progresses in his design, printing and testing, reports of listening, and subjective evaluations, will become a significant part of this thread

[Pete Schumacher]

Were these conical at the throat, or curved? If curved what was the radius?

circular arc profiles tangent at the mouth on both. The radius for the 2" deep, 7" wide arc can be calculated since only one is possible, same for the 8" wide, 1" deep guide. I don't recall the radii off hand.

[augerpro]

That's enough for me to figure it out, thanks Pete!