3D printed waveguides

[augerpro]

Have you considered an opening which sits above the dome and occludes the surround? The guide we made for the T-Labs tweeters performs exceptionally well that way. Not a hint of top octave wiggles.

I had tried that with the 3" Tangband dome and it had a pretty bad impact on the response. I've found that others *usually* have the same issue, so I didn't do it here. I'll make a part though and see for the heck of it. I'm going to do another batch with various tweaks to tease out what is doing what and where. And a few phase shield tweaks.

Did you try any versions of that with the throat outside the surround? How deep is that? Good looking speaker, bet they sound great!

[augerpro]

My change was going from a round opening to a square opening. As the null at the top end is result destructive interference, I think working with the edges or center is a way to change the summation. My theory with dome tweeters is we are trying to convert a curved wavefront to a plane wave front.

As I understand it, the horn is always wanting to output a spherical wavefront. This is tricky since CD's *ideally* output a plane wave that must then be bent. But with dome tweeters, they are already making (some form of) a spherical wave, and so from that perspective the simple cone is perfect to maintain that. That's the ideal anyway. Of course it has an issue at the mouth, and once you start making a roundover there on a small guide you quickly become a curved guide.

So I got a few more designs to tease some of this out. And the throat angle may be the critical factor here, as we go deeper, perhaps the shallower flare of the non-tangent versions will be the better performer because it can meet the dome at a similar angle as the tangent ones currently do? Anyway that is why I want to spend some time here at 5" and try different things, even if I'm pretty sure it won't work.

[Pete Schumacher]

I had tried that with the 3" Tangband dome and it had a pretty bad impact on the response. I've found that others *usually* have the same issue, so I didn't do it here. I'll make a part though and see for the heck of it. I'm going to do another batch with various tweaks to tease out what is doing what and where. And a few phase shield tweaks.

Did you try any versions of that with the throat outside the surround? How deep is that? Good looking speaker, bet they sound great!

Beryllium T-labs (28mm) paired with custom Audio Technology Flexunit 8" woofer. It does sound incredible. What a combo!

That guide is 2" deep, 8.5" diameter with a 1" throat that was cupped on the back side to allow for a bit of clearance for the dome movement. That opened up the hole to almost 28mm diameter.

The original guide we tried was for the RS28A/F that basically continued the opening of the tweeter faceplate (1.625") and it too showed no wiggles in the top octave. Because we didn't remove the faceplate, the dome was recessed and didn't protrude into the guide opening.

[TN Allen]

Pete,

I've been thinking about leaving space between the dome and throat, and keeping the throat diameter smaller than the surround may offer some benefits as well, but haven't tried it yet. What would you suggest for a contour/cross section. I can mill or turn something to throw into the test mix.

[jhollander]

On my to test list is a card stock washer to delay/ shape the tweeter edge response. Second test would be felt treatment on the beginning edges of the WG. If you try any of those please share the results.

Pete's note is interesting as the edge diameter of the RS28A is more like 1.875". Seems like smaller diameter would fit the results above.

[TN Allen]

I'm willing, within reason, to mill or turn test guides to test design ideas, good bad or otherwise. I don't test the guides I produce, other than listening, but am happy to produce other's ideas for testing.

Send me a CAD file or sketch, and I'll see what I can do.

The RS 28 is tricky if the face plate is removed. The surround/throat diameter I've used is 1.475", but it can be difficult to re-center the voice coil, resulting in unpleasant sounds. The Usher 9950 (?) is also tricky. If someone wants to test these drivers in a guide without the face plate, I'll also help with this.

[Pete Schumacher]

I'm willing, within reason, to mill or turn test guides to test design ideas, good bad or otherwise. I don't test the guides I produce, other than listening, but am happy to produce other's ideas for testing.

Send me a CAD file or sketch, and I'll see what I can do.

The RS 28 is tricky if the face plate is removed. The surround/throat diameter I've used is 1.475", but it can be difficult to re-center the voice coil, resulting in unpleasant sounds. The Usher 9950 (?) is also tricky. If someone wants to test these drivers in a guide without the face plate, I'll also help with this.

After testing a bunch of guides made by DanP, I found the circular profile provided the smoothest response for both on and off axis. And they worked great with the faceplate of the RS28 (and Usher and SEAS 27TBFC) left in place.

One other thing to consider is the opening angle. If it's less than 90 degrees at the mouth the high frequency response seems to suffer.

[TN Allen]

Over the past few years I've used the RS28 and 9950 in various combinations of drivers, always with a guide similar to that described by Pete, ~8" diameter X 2-2.25" deep, but without the face plate. The sound is robust and clear, very satisfying. I'll need to try the driver with the face plate in place.

It might be beneficial to do a set of guides for the RS28 to test various configurations, for example with and without the face plate, and with and without the surround occluded. Perhaps in doing this we can methodically provide a repeatable baseline data set with a well known tweeter. New guide contours could then be produced to test new ideas and expand the baseline data. Perhaps there is a better tweeter for this, the RS 28 seems popular though and has a fairly broad response range.

[augerpro]

I've designed a version that will overhang the surround as Pete mentioned, based off the 1" tangent mouth waveguide. Also included dimples to index the tweeter for easy alignment:



I've also made smaller phase shields since that seemed to be better experimenting last night. Also going to try one scalloped:



Any other suggestions?

Pete when you say "One other thing to consider is the opening angle. If it's less than 90 degrees at the mouth the high frequency response seems to suffer." you mean the flare (mouth) meeting the baffle tangent, correct?

[TN Allen]

Brandon,

Can you please show a cross section? That would be useful.

Also, it might be a good idea to consider that many people on the forum can probably look at .dxf and .iges files, perhaps you can post these on a "Drop Box" like service with a link. One of the short comings of previous WG discussions has been accurate detailed geometry. It would be good if we can all look at the same accurate CAD files.

Thanks

[augerpro]

Let me know if this works: https://www.dropbox.com/s/y6s2b4lozv...25_17.dxf?dl=0

[TN Allen]

Thank you, the CAD geometry and test results may help those interested to evaluate correlations between the two more objectively.

[augerpro]

Does that .dxf file have what you want?

[TN Allen]

I'll look tomorrow. If it's the 3D geometry, I can find what I need. Although, I think in the past we exchanged .iges files too. Perhaps others will look as well. Thanks for posting the file.

[augerpro]

Playing some more with the phase shield. This is on the .75" deep, tangent mouth guide. Let me know if the bolder colors are easier to read or not.

First the original version, with .47" wide phase shield:


Widdling the PS down to roughly .38":


Smaller yet, .25" wide:


And no phase shield: