3D printed waveguides

[ibthal79]

3D printing is a disruptive technology, offering the inherent capabilities for creating truly arbitrary 3D structures, with low manufacturing costs associated with low volume production runs. This paper provides an overview of the current progress in 3D printing of metal-pipe rectangular waveguide (MPRWG) components, from 10 GHz to 1 THz, at Imperial College London. First, measurements performed at the UK National Physical Laboratory demonstrate that 3D printed MPRWG performance is comparable to standard commercial waveguides at X-band and W-band. Then, a fully 3D printed X-band dielectric flap tuneable phase shifter and W-band 6th-order inductive iris bandpass filter are demonstrated experimentally. Finally

[TN Allen]

"3D printing is a disruptive technology, offering the inherent capabilities for creating truly arbitrary 3D structures, ..."

Is the above a quote? If so, could you please provide a link?

Most technology is disruptive, especially initially. 3D printing is only one new tool, and not that new. It developed out of Stereolithography, which has been around at least 2 decades. The .stl file format is from Stereolithography, as far as I know.

From what I've seen 3D printing needs considerably more development before it supplants subtractive machining in the range of materials that can be machined, accuracy and efficiency. There is currently a lot of hype regarding 3D printing.

Regarding "truly arbitrary 3D structures", this may be another way of saying non-geometric primative shapes. The real basis for this is CAD, and advances in motion control technology. CAD and CNC machines truly have revolutionized manufacturing. 3D printing has a ways to go before the same can accurately be said for it. Claiming it is "disruptive" is an interesting advertising ploy.

[augerpro]

Yeah I'll have to look into the Revel Ultima2 measurements. Looks a bit shallow, with large radius flare. That combination didn't work for me earlier BUT that was when I was still using round waveguides and didn't have the throat and phase shield figured out. Don't be surprised if you find measurements and they show problems, my own SB26 guide is better IMHO than the one Revel uses in their M105/106 speakers. I linked to their paper earlier in this thread.

I can whip up some designs for you to print and measure, just let me know.

[silverD]

"3D printing is a disruptive technology, offering the inherent capabilities for creating truly arbitrary 3D structures, ..."

Is the above a quote? If so, could you please provide a link?

I think this is a bot....look at the other posts from this user.

[augerpro]

Getting back on topic a bit. Looking at the results for the larger dome Satori and SB tweeters is disappointing; they have such great sensitivity and power handling for domes. I see Revel uses similar large surround domes in their flagship Ultima2 line. Other than the throat starting outside and behind the surround, I don't see any glaring differences. If Revel uses it in their best speaker, surely it wouldn't have glaring top octave issues?

From the review at Stereophile Revel does a decent job keeping the top octave fairly flat on axis, off axis not so pretty:

[bwaslo]

The Prusa can't print it (bed not big enough). Needs to be 300mm x 300mm and about the same in height. I used an FT-5, with lots of mods (most importantly, a Volcano hot end to handle the larger nozzle size).

If you want to make an ever BIGGER horn, I designed a "stub", which provides the shape for the throat and ports for the midrange injections. It's intended to be used with plywood wall extensions, with woofers injection through those. How big? As big as you can go. (I never built it, though I've printed several of the stubs -- I woke up and realized that I have no where to use or to even store anything like a 5 ft horn!).

If anyone wants the stl file for that, it's at: http://www.libinst.com/SynergyDIY/StubV6/V6%20stub.stl. If you do it, make sure to post photos (of distressed neighbors as well as of the speakers).

Print time itself really isn't all that big of an issue (it's the 3D printer's time, not yours!). But the longer a print goes and the larger it is, the more the likelihood of something going wrong during the print.

[jhollander]

That is an ingenious idea Bill. If someone comes up with some curved terminus transitions I may have to build another horn.

[Mandroid]

Finally got the waveguide print from 3dhubs. This was in black PETG. Print quality on the front is serviceable, layer adhesion isn't great in spots, but some finish work could probably clean it up well enough. The rear is quite a bit worse. The layers are very inconsistent with much material missing. The shim layer at the throat is not uniform. The tweeter fits well on the alignment pins. The phase plug looks okay. It's slightly flexible, but seems well attached. Not sure if it could cause a resonance, though.

The rear flanges and throat seal in particular are going to be issues with this one. The SLS prototypes augerpro is working with are of vastly superior quality. With a bit of wasted material and time, I hope I can sort out the settings on my own printer to produce a viable component.

[augerpro]

That rear seal is critical. And no one wants to spend a lot of time finishing these to be presentable. Who was the actual vendor you used?

[Mandroid]

3dhubs recently dropped all of their independent print hubs. Hobbyists can no longer use the site to get print work to offset the cost of their printer. There is no option to choose what vendor performs the print any more. Mine came from a fulfillment center in Georgia.

I agree that the part I received is unusable, and will be doing all my own printing now.

[augerpro]

Huh, interesting. What printer are you going to use?

[Mandroid]

The Folgertech FT-5 R2. Would have gone with prusa, but wanted the larger print area to try more advanced baffle shapes and have ability to print bwaslo's synergy horn.

[TN Allen]

STL can't be converted to anything, unfortunately. It's just a collection of triangles.

Dan

I remembered this brief discussion when I was trying to import a .stl file in order to generate a tool path for the design. The .stl file comes in as a shell, and other than scaling it up or down, nothing else could be done. However, I spoke with a man who suggested an option in Key Creator for importing a .stl file as a solid or surface. That worked well. I mention this because I suspect other CAD programs may have a similar option, which could be quite useful if no other format is available.

[augerpro]

More waveguides results! I just changed my soundcard so I can now measure out to 48khz and capture these metal dome breakups a bit better. I'll keep updating over the next few days as I have a lot of waveguides. Up first are the 5" waveguides.

I knew the CSS LD22 could measure a lot better with a smaller throat, and these results definitely prove that out:



Looks damn good!

Next are the RST28's. First the fabric dome with no phase shield:



Now the RST28A with the standard .25" diameter phase shield I've been using, it is about 1/8" from the dome:



Holy breakup! I think there is room to play with the PS to hopefully get a null to land on the breakup. I won't post all of the variations I tried, but I moved the PS closer and farther away, used a .47" dia PS, added an outer ring and played with ring size. The ring looks to be an improvement when comboed with the .25" dia PS, not so much when comboed with the .47" dia PS though it was the best result of any other .47" variant. Here are the best three results of the experiements I did:




And also the RST28A without a phase shield:



Will get the SB21 and SB26's up tomorrow, I now have three variants of the SB26: aluminum, ceramic/anodized Al, and fabric. I also have some 4" and 8" waveguides.

[jhollander]

Thank you for sharing that is great work