3D printed horn experiments

I've been getting my feet wet in the world of horns and wanted to share my initial experiments here- I hope any readers out there will find it mildly interesting and perhaps be able to give me some advice as well! I would describe what I'm doing as "bumbling around" and entirely welcome any questions or pointers people might be willing to put forth.

Horns actually first caught my attention as potential woodworking projects... I'm pretty new to the hi-fi hobby and a lot of the enjoyment I've gotten from it has been with tinkering around with different components, making parts and pieces and whatnot. I really like the aesthetic of the vintage conical wooden horns but alas, I live in a tiny apartment with no space for such things. Researching the wooden horns led me into the realm of 3D printed horns, and I happen to work a job that involves working with large format 3D printers so it was a pretty quick change of direction to go from wood to plastic.

I'm sure none of you will be surprised to hear that I got to talking with Trieu on here, and reading a bunch of his threads with his various experiments with modeling and printing different horns. He has been an invaluable resource introducing me to the world of horns and we came to an arrangement allowing me to print some horns using one of his expertly modeled CAD files- the WP12024. If you want to read up more on other's experiences with this particular horn you can find it in thread "Matrix multicell horns..."

A few details for anybody who dabbles in 3D printing... I work developing and testing large format, pellet-driven extruder heads. The prototype horn that I printed to test with my setup was printed with PETG and a 2.5mm nozzle, 2 mm layer height. The whole assembly is actually 4 modular pieces, the six cells are split into three pieces each having two cells, and a throat adaptor that connects to all of them. The print quality is somewhat reduced with a 2.5mm nozzle but the prints complete very quickly so I chose to go this route for the initial prototype and then print "keepers" with a smaller (1.5mm) nozzle. I also skimmed the finished printed horns with a coat of epoxy as an experimental resonance damper / base for applying wood veneer. With the drivers mounted they are impressively heavy.

The adaptor is sized for drivers with 1" throat openings- I was fortunate to catch a pair of 802g drivers (blessed and cursed to have entered this hobby as an Altec enthusiast) for sale here on the Emporium which I grabbed to play around with these horns. Future plans are to try to source some woofers (414 is most likely candidate) to build my own cabinets to pair with a finished set of horns. At the moment for trialing the horns I'm just running them side-by-side with my Altec Boleros to fill in the bass. The 406 is the little brother to the 414 so it seems to have a decent logic to it, though for the purpose of my test print audition I've just had them hooked up A+B on my Sansui 7070 with a cheap 800 hz high pass filter to protect the 802g drivers... somewhat uncouth. I'd like to play around with making my own crossovers for the 802g drivers once I've settled on some final bass woofers to pair with them and would happily take any and all advice in this realm. I'm confident with my soldering skills but have only basic theoretical knowledge of making crossovers.

Anyway that is a lot of words... here is my listening impressions so far! Keep in mind this is a fairly unrefined audition setup- I only printed one horn so far so have been doing my listening in mono, and I need to at least wire in an independent volume control for the horn because it outperforms the Bolero cabinet in the sensitivity department.

The Wp12024 with the 802g definitely provides a very detailed listening experience. The first album I put on was "Not Our First Goat Rodeo" w/ Chris Thile, Yo-yo Ma, etc. which has become a recent favorite reference for system tweaks. I felt like I could hear more "textures" than before, small details like the sound of the bow on the violin strings. On crescendos the violin sounded harsh, which from what I gather is sort of a characteristic of horns in general. Of course, everything is relative to what we get used to and if I listen to an entire album with the horn included then turn it off my Boleros on their own sound much more laidback or "polite" but lacking a certain liveliness.... if I'm listening to just the Boleros when I first kick on the horn the immediate impression is of harshness. Proper balancing would hopefully create the middle ground between these two flavors.

I was surprised to find that my expectations of what albums would sound good on the horn were totally incorrect. I thought I would like hearing jazz, bluegrass, classical type stuff on them - but I ended up really enjoying stuff like Courtney Barnett, Waxahatchee, and of course I had to play Quadrophenia. Nothing like a french horn on a horn! I was also pleasantly surprised to find that while feminine voices sounded as good as I expected, the deep rumble of Gregg Brown's vocals also gained a little extra something coming out of the WP12024.

The horn ended up being larger than I imagined and probably not the best pairing for my small listening room where I am usually sitting quite close to the speakers. It doesn't really come into its own until you put a certain amount of volume into it, but there isn't much of an in between from quiet to blow your ears off - I have yet to put more than 1/2 a watt of volume to them which definitely gets you to "concert for the neighbors". Until I have a larger place to live I might have to keep some sort of A/B setup where I can selectively play the horns at appropriate times because if they were on all the time my wife might leave me.

I suppose this thread will serve as a sort of project diary moving forward - I'm not sure if my next move will be to print a second horn so I can listen to the WP12024 in stereo or wait until I've sourced a pair of woofers to make some cabinets with (Eyeing a pair of 414z on auction right now). I will likely do some more experimenting with variables in the signal path with the current test setup as well, first order being some way to pad down the volume on the horn. I've searched around here and found some schematics for crossovers for the 802s, but it isn't clear to me if I could essentially just use the HF half of the two way on its own for the horn. Any advice here would be greatly appreciated, and let me know if you have any questions about the prints!
 
Very cool! I’m glad you’re enjoying the 802-8G’s! Not sure what your listening positioning is, or if it even would make a difference but I would put the horns more pointed at somewhat ear level. Anyway, I know you’re still in the trying things out phase anyway. Looking forward to seeing what else you come up with as it’s sure to be cool. Any ideas for crossovers?

best,

jonas
 
Very cool! I’m glad you’re enjoying the 802-8G’s! Not sure what your listening positioning is, or if it even would make a difference but I would put the horns more pointed at somewhat ear level. Anyway, I know you’re still in the trying things out phase anyway. Looking forward to seeing what else you come up with as it’s sure to be cool. Any ideas for crossovers?

best,

jonas

Hey Jonas! Yes very much enjoying experimenting with the 802s, thanks again for the connection!

I think the perspective in the photos is maybe a bit deceptive, while seated they are actually pretty dead on to ear level (perhaps eye level is most accurate) though they are tilted ever so slightly upwards due to the back of the horn overhanging the back of the cabinet at the moment. As you mentioned, they are firmly in the experimental phase right now- I'm actually going to print one other style of horn with a slightly smaller footprint and see how they compare.

Just won an auction yesterday for some 414z drivers so throwing a cabinet build into the mix as well to sub in for the Boleros.

Still in the "research" phase of getting some proper crossovers put together, got some kind advice and direction on the topic from @jrocker

I'll keep updating this thread as things come together, I'll surely be tinkering away on one piece or another.
 
Everything is hard until you’ve done it.

This one is not that bad.... Free high frequency extension and better imaging.
I have a pair of mismatched 802D's. One came with the bug screen already removed. Would have been interesting to listen back and for to see if I noticed a difference, but I ended up removing the bug screen on the second one. I put NOS diaphragms in the 802D. I did compare them to a pair of 802-8G's (which still had the bug screens) and the 802D's do not have as much high frequency extension as the 802-8G's, even with the bug screen removed in the 802D's. That is not to say that the 802-8G wouldn't benefit from removing the bug screens, that I can't say, but I feel they are already doing well in terms of high frequency extension.
 
Excellent! This should be fun and with some experience the past few years in the CNC realm myself, I understand how much time and effort goes into the initial phases of design modeling and set-up. Usually followed by adjustments to fix the one or two off settings or values.
3D printed horns I've seen have not been inexpensive so it will be interesting to get your impression of the costs associated with this overall and how that might translate to what could be made available.
I am no expert but I would not underestimate the impact that those rough printing finishes (2.5mm nozzle for draft purposes) might have on the sound quality. Others have fretted over seams in the throat.

What I would like to see frankly with some of these types of attempts is production of quality horns of a scale more conducive to an actual home environment. The 414 is a great start in that it being a 12" driver and not 15. More reasonably sized horns would be my preference. Plus this should help overall on production costs in both the horn printing as well as cabinet builds with the prices of wood going crazy. (MDF 4 x 8 3/4" panels here jumped from $31.99 to $39.99 in the last few weeks at Home Depot). Someone needs to tell those Canadians to get things under control up there and start producing more lumber!
 
I have a pair of mismatched 802D's. One came with the bug screen already removed. Would have been interesting to listen back and for to see if I noticed a difference, but I ended up removing the bug screen on the second one. I put NOS diaphragms in the 802D. I did compare them to a pair of 802-8G's (which still had the bug screens) and the 802D's do not have as much high frequency extension as the 802-8G's, even with the bug screen removed in the 802D's. That is not to say that the 802-8G wouldn't benefit from removing the bug screens, that I can't say, but I feel they are already doing well in terms of high frequency extension.
I'll make sure to try an A/B listening halfway through the process and see what I can hear, and report back!
 
Excellent! This should be fun and with some experience the past few years in the CNC realm myself, I understand how much time and effort goes into the initial phases of design modeling and set-up. Usually followed by adjustments to fix the one or two off settings or values.
3D printed horns I've seen have not been inexpensive so it will be interesting to get your impression of the costs associated with this overall and how that might translate to what could be made available.
I am no expert but I would not underestimate the impact that those rough printing finishes (2.5mm nozzle for draft purposes) might have on the sound quality. Others have fretted over seams in the throat.

What I would like to see frankly with some of these types of attempts is production of quality horns of a scale more conducive to an actual home environment. The 414 is a great start in that it being a 12" driver and not 15. More reasonably sized horns would be my preference. Plus this should help overall on production costs in both the horn printing as well as cabinet builds with the prices of wood going crazy. (MDF 4 x 8 3/4" panels here jumped from $31.99 to $39.99 in the last few weeks at Home Depot). Someone needs to tell those Canadians to get things under control up there and start producing more lumber!

Yes you are right that there is a fair amount of "behind-the-scenes" work that goes into any of the CNC type projects that adds up before you get to the point where it could be considered automated and even then it's a little bit more then just hitting the print button... working in the 3D printing industry I get a lot of questions about it and it's amazing how often people think that you can just send a photograph to a printer and have it print what you want... Seriously!

Want to reiterate just in case people jump into this thread midway that I didn't do any of the modeling for these horns, which I'd say is the hardest part of the process - these were modeled by @Tsingtao_1903.

Of course safe to say I'm with you on being curious to explore the world of "domestically scaled" horns, since I don't as of this moment have a large, dedicated listening space. I know Trieu has played around with some of these smaller horn models.

As to the influence of "texture" on sound quality I'm sure I'm even less of an expert than you 😄 though I have heard it said that a subtle texture such as created by a smaller nozzle (say .6 or .8) is actually desirable for diffusing reflections and smoothing out the sound. I'm no acoustic engineer... I think that the texture of the 2.5 is undesirable enough just on an aesthetic basis that they shouldn't be used other than as a quick draft prototype. I will say I haven't noticed any sonic difference between the 2.5 and 1.45, would love to be able to compare any of the 3d printed horns side by side with a cast metal original but alas, I have not the means. I took pictures of the inside of the horns to give an idea of the texture of both.

Will follow up with you on thoughts on cost... material cost i.e. just the bulk polymer is not inconsiderable and for anyone that doesn't have a large format printer these prints take a long time to complete... I think if you are having someone else print your horns they are going to charge a lot for this reason.
 

Attachments

  • IMG_20210316_175917.jpg
    IMG_20210316_175917.jpg
    495.1 KB · Views: 46
  • IMG_20210316_175936.jpg
    IMG_20210316_175936.jpg
    482.9 KB · Views: 45
Haven't had a chance to post any updates for about a month but have still been chipping away on this project piece by piece. I printed a smaller pair of horns (KS12024 model by @Tsingtao_1903 ) which I've been listening to now for a few weeks, still just set up in a temporary fashion A + B with my Boleros with the quick and dirty pass filter + l-pad to bring down the efficiency. It was interesting to be able to do a side by side comparison with the initial larger horns and hear the sonic differences. I'm preferential to the KS though the WP had some nice qualities as well. KS to my ear sounds more detailed and crisp with the instrumental textures more forward while the WP gave an impression of the sound coming from a larger room, more space and air and a bit more relaxed sounding.

I've been getting crossover advice from @jrocker who is extremely generous with his knowledge, and I ordered some Baltic birch ply last week that should come in this week so I can finally build some cabinets, which I sourced 414z drivers for. I've been swinging the ol' carpenter's hammer on the weekends and haven't had any time to myself so extremely excited to start building next weekend!

Should have some juicy progress photos next week, my target volume is 8 cubic ft... hoping they don't look too hulking. Trying to follow recommendations in the old Altec literature.
 
To my great disappointment, my plywood was not delivered on time this week and I will have to wait until next weekend to start building my speaker cabinets. Since I was hankering to make something, I threw together a little stand for the turntable/amps. Nothing fancy, I actually made it entirely out of scraps that were at the shop, and I even dug up some threaded leveling feet to compensate for the roley-poley floors in my apartment. My welded frames never come out quite square but I guess that just means I need to practice more!

Did a little bit of crossover tinkering as well today, should be in good shape to have a starting point for when the 414s are ready to rock! IMG_20210425_160549.jpg
 
After my plywood finally came in this week I was able to get cracking on my 414 bass reflex cabinets today! 🎊🎊🎊

(Mandatory complaint about the obscene price of plywood)

Didn't quite get all the construction wrapped up today but I'm in good shape to be listening to them by the end of tomorrow! 5/6 sides glued and assembled, just need to cut my holes for the 414 and the ports in the front baffles, screw some bracing in, and pop the fronts on. Oh and sand everything... finish... etc. etc.

Then there's frames for grilles and something fancy looking to mount the horns to but those details are being kicked further down the line...

Here's the only process photo I thought to take today. Went with splined edge miters for 4 sides of the box, and rabbets to let in the back panel flush, which is permanently glued in place.

IMG_20210501_142304.jpg

More photos to come tomorrow!
 
I’ve heard about construction materials going up, what percentage did plywood jump up? I’m looking to buy some walnut 3/4 ply to build an “entertainment center”/cabinet for all my gear.
Excited for ya too!
 
Back
Top