3D printed waveguides

[augerpro]

Will probably play with throat size variations to, but I'm pretty sure this will end up being the right fit:

[TN Allen]

It's good to see you're back at this project. Any chance you might say more specifically what the printing process is for the new guides, and what they cost?

[augerpro]

I'd have to look it up again. I guess similar to a SLA printer, but with powdered nylon instead of resin.

[augerpro]

Ok here are the first results. This is a bit of a data dump, haven't really looked at the results myself in detail yet. A little about my measurement setup: I use MLS measurements with 6ms of clean signal for a quasi-anechoic response, no smoothing, and a calibrated mic. Pictured is the baffle, I don't have my big baffle anymore, so I optimized this the best I could, it does appear that under about 4khz the baffle is having a strong impact. Now IF the waveguides are doing there job, this effect should go away. This first batch is the nylon sintered ones, with phase shield. I'll be posting each update at my Soma Sonus site, so that will probably be easier to get updates on than this thread.

All waveguides are 5" diameter mouth. The rest of the info is listed below.


Flat baffle for reference, 0-10-20-30-40-50:


3/4" deep, tangent mouth, 0-10-20-30-40-50:


0-30-60:


profile:


3/4" deep, non-tangent mouth, 0-10-20-30-40-50:


0-30-60:


profile:


1" deep, tangent mouth, 0-10-20-30-40-50:


0-30-60:


profile:


1" deep, non-tangent mouth, 0-10-20-30-40-50:


0-30-60:


profile:


Next I'll see what it is like without phase shields, play with phase shield placement a bit, lower radius non-tangent mouth versions possibly even a straight cone, and a roundover on the non-tangent versions.

Any other measurements you'd like to see before I cut the phase shields off?

[300Z]

Looking good and glad you'll be keeping the future updates on your website too.

[dlneubec]

Hi Brandon,

Glad to see you are still working at this project. I have not been following closely, but just happened to check this update. Your results look very encouraging. It seems that the non-tangent mouth is creating a bit of a depressed shelf. Eyeball test favors the tangent results to me.

Here would be my suggestions, FWIW. I think it might be helpful to do a sum of your 5 measurements and post that above the others. This will give somewhat more of a clue as to what the overall power response will look like in a single line. I would save the sum separately for each guide, so you can recall them and compare the sum/power response of each guide to one another. This might make it easier to see the differences in how they might actually sound because you are comparing just one line for each guide and then correlate that to what you hear when listening. You might eventually find a sum shape that seems to produce the best result.

I suspect you will find a significant and measurable difference based upon phase shield size and placement. Closeness to the dome is likely to impact a different portion of the top octave, just based upon different wavelengths involved. I've even wondered if a horizontally offset phase shield might be result in a significantly different response, in that it would not shield at the exact same frequency off the tip of the dome, sort of similar to an offset tweeter in a baffle allowing the diffraction frequencies to spread out, somewhat, at least at 0 degrees off axis. I'd be curious to see what that would do. I've also wondered what a tiny cone shaped phase shield might do to the response, as opposed to other shapes and/or no shield at all.

Anyway, great stuff!

[scottvalentin]

Brandon great job and thanks for sharing those results. Insightful feedback Dan, thanks for providing your input.

[augerpro]

Thanks for the feedback Dan! It's funny you mention summing the responses, I'd actually shifted to that sort of thing in my speaker designs. Inspired by Sean Olive's paper I now do Direct Sound (0+10) and Early Reflection (30+40+50+60). As you said, it has really brought a better perspective to the response, especially the ER portion. I had to play with delay so the phase differences wouldn't cause incorrect cancellation, but now I see Soundeasy V22 has a power response and directivity calculator based off the measured off-axis response (I believe, haven't used it yet). So hopefully they will add as minimum phase.

Yeah I suspect the phase shield will be quite a difference, I'm going to try bringing it out about 1/32", and also making it smaller. The phase plug shape is a good idea, once I nail down a "final" candidate I'll try that. Question is Scanspeak shape (concave) or Vifa shape (convex)?

[dlneubec]

I rarely design with direct axis measurements. Haven't for a number of years. I usually do 0-60 in 15º increments, then sum. I compare the sum to the 5 single measurements and choose the single that best matches the summed response. Sometimes none are quite right and I have to take an intermediate, like 22.5º to get a good match with the sum. The measurement that best matches the sum is generally where I start with crossover design, I have found that there is much less need for "voicing" adjustments than when I started with on axis measurements only.

Agreed, you do have to take care with your setup to be sure your distance or delay is exact or your phase is matched as perfectly as possible at each angle to avoid cancellations. Was about to upgrade to V22 myself, but have not gotten around to it yet.

[augerpro]

Ok Dan just for you:

With a little cone made from blue tack:



Causes a big null to move lower, but also filled in the mid-treble dip. Interesting.

No phase shield at all:


And last, PS reattached with some blue tack, so it was just a hair farther forward:


Right now the PS is approx. 1/16" from the dome. I did some measurements with it 1/8" from the dome just now, but forgot to save the results. Anyway a big null dropped down in the 10-20khz, area so it appears closer is better. The factory spacing was about 3/32".

[jhollander]

For my synergy horn I stacked felt round pieces on the phase shield/ mesh of my RS28A which smoothed the response a bit, although stack of 6 did deepen the 18K null. The XT25 modeled the best with no null. It was interesting that while I'm using a conical horn, the response improved with a larger initial tweeter opening.

The response above looks similar to my RS28A in a deep 45 degree chamfer (different project). Note the the RS28A has a built in phase shield.

[augerpro]

It was interesting that while I'm using a conical horn, the response improved with a larger initial tweeter opening.

How big relative to the surround?

Definitely some room to play with the phase shield. I've made the blue cone above smaller and smaller until it was just a little dome, none were as good as original little disc. Also making the disc bigger was worse. Smaller smoothed out the hump/dip in the top octave of the 1" tangent mouth, although I cut it with an exacto knife so it wasn't round either, not sure what made more difference. Worth investigating more for sure.

[Pete Schumacher]

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.

[dlneubec]

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 know DaveP had started doing that with a number of his guide designs and found it improved the response, at least for some tweeters. He would create a faceplate his heat pressed guides fit into and used something like a small cove bit on the back of the face plate so it essentially hovered over the surround.

[jhollander]

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.