Hi all,
Thinking about build one of these, a 12 faced omni.
First driver that comes to mind is 3FE25,
because I already have 4, and 8 more would only cost me another hundred dollars.
I'd like to get reasonably flat freq response from 100Hz, to as high as possible, without excessive low end boost.
For any of you folks who have contemplated a dodec, any driver recommendations come to mind ?
And did you consider porting them?
thx 🙂
Thinking about build one of these, a 12 faced omni.
First driver that comes to mind is 3FE25,
because I already have 4, and 8 more would only cost me another hundred dollars.
I'd like to get reasonably flat freq response from 100Hz, to as high as possible, without excessive low end boost.
For any of you folks who have contemplated a dodec, any driver recommendations come to mind ?
And did you consider porting them?
thx 🙂
You aren’t building a pair?
The optimum miniOnken for 3F25 is 2.5 litres, you’d need at least 32 litres and a dodecahedron encloses close to the minumum possile volume for the external size.
I would suggest sealed with helper woofers. As big as you can live with. EQ may be benficial.
dave
The optimum miniOnken for 3F25 is 2.5 litres, you’d need at least 32 litres and a dodecahedron encloses close to the minumum possile volume for the external size.
I would suggest sealed with helper woofers. As big as you can live with. EQ may be benficial.
dave
Thanks Dave,
I never build a pair at outset.
For me, first try is always a proto to learn from. When i feel i've got it right, then I build two.
When you say, 'as big as i can live with', were you talking the dodec volume or the woofers size?
I'm guessing woofers, because sealed volume for 3fe25 doesn't seem to model any better with volume past a couple of liters or so.
If woofers, i've got that covered and then some.
If for the 3fe25, i'm more inclined to vented than sealed.
Here's a hornresp input screen for a single driver vented.
Yay or nay?
I never build a pair at outset.
For me, first try is always a proto to learn from. When i feel i've got it right, then I build two.
When you say, 'as big as i can live with', were you talking the dodec volume or the woofers size?
I'm guessing woofers, because sealed volume for 3fe25 doesn't seem to model any better with volume past a couple of liters or so.
If woofers, i've got that covered and then some.
If for the 3fe25, i'm more inclined to vented than sealed.
Here's a hornresp input screen for a single driver vented.
Yay or nay?
Attachments
Perhaps TC7FD? Nice response. 12 would be 1.2 cubic ft sphere - not too big. Vented alignment is 86Hz according to factory. Crazy sensitivity if wired 4series/3parallel circa 5ohms.
Peerless by Tymphany TC7FD00-04 2-1/2" Full Range Paper Cone Woofer 4 Ohm
Only $8.17ea
Peerless by Tymphany TC7FD00-04 2-1/2" Full Range Paper Cone Woofer 4 Ohm
Only $8.17ea
Thx xrk,
Popular little bugger...out of stock PE and Madisound, expected in May by PE.
Any reason to prefer it over a TC9?
Cause I also have 4 leftover TC9's from line array projects.
Been trying to decided between them and the 3fe25's...
need 8 of whichever, or 12 of something entirely different if another compelling driver pops up
Popular little bugger...out of stock PE and Madisound, expected in May by PE.
Any reason to prefer it over a TC9?
Cause I also have 4 leftover TC9's from line array projects.
Been trying to decided between them and the 3fe25's...
need 8 of whichever, or 12 of something entirely different if another compelling driver pops up
Only that they are less expensive and require less volume. Not sure if 12 TC9’s can reach 86Hz tuning in a 1.2 cubic foot box. But if you can grow your volume, TC9 is fine. The 3FE25 is a great driver and should work well. The 3FE22 will let you make a smaller box but is 2x the price.
If you can fit them, you should have plenty of the TC9's for prototyping right?
They may be lacking for Dave, but they do (way) more than fine for most of us that aren't afraid of a little EQ.
They may be lacking for Dave, but they do (way) more than fine for most of us that aren't afraid of a little EQ.
I've modeled some dodecahedrons, and it's really really difficult to get any kind of high frequency extension.
For instance, at 10khz, if you have two wavefronts that are 1.7cm out-of-phase, you'll get a null.
And as you can imagine, with a twelve sided speaker, you're going to have pathlength differences of *significantly* more than 1.7cm.
A ring like the Horbach speaker could be interesting...
To make the rings in the Horbach speaker, you'd want to shade most of the drivers in the ring, so that most of the output is coming from the driver that's closest to you. Imagine a Keele CBT, but ring shaped.
For instance, at 10khz, if you have two wavefronts that are 1.7cm out-of-phase, you'll get a null.
And as you can imagine, with a twelve sided speaker, you're going to have pathlength differences of *significantly* more than 1.7cm.
A ring like the Horbach speaker could be interesting...
To make the rings in the Horbach speaker, you'd want to shade most of the drivers in the ring, so that most of the output is coming from the driver that's closest to you. Imagine a Keele CBT, but ring shaped.
Mark, did you see prior experiments from barleywater?
https://www.diyaudio.com/forums/multi-way/236290-uniform-directivity-71.html#post3643935
This is more in-line with what Patrick suggests, without the needed DSP to preserve the top end, I believe nothing much came from that experiment. The top end is gone though when run full range, as seen in the measurements.
https://www.diyaudio.com/forums/multi-way/236290-uniform-directivity-71.html#post3643935
This is more in-line with what Patrick suggests, without the needed DSP to preserve the top end, I believe nothing much came from that experiment. The top end is gone though when run full range, as seen in the measurements.
That's a great thread about directivity, and in general as well. Thx.
The biggest motivator behind the dodec project is for room acoustic measurements, but i sure as heck hope it sounds decent too.
Maybe I'm missing something, but i've been thinking a 3" driver will have ka = 6 at 10kHz, for about a 60 degree beamwidth.
The dodec has a 66 deg splay angle between faces, so I'm thinking combing might not be a big deal from 10kHz up??
And be more likely to be a big deal below 10kHz, down to maybe where c2c spacing gets relatively tight?
Swicthing gears to 'what driver.'...... a modeling question for you guys...
In trying to choose between drivers, I've been looking at each in hornresp, sealed and vented. But only modeling a single driver
I've been assuming i can just multiply volume by 12, for the dodec size. Is this ok?
But how about ports? 12 little ones, or combine into fewer and larger?
Doesn't seem like port area and length can scale linearly (like maybe volume can)?
Hope those questions made sense...
Thx, (from a guy who tries to sim as little as possible Lol)
The biggest motivator behind the dodec project is for room acoustic measurements, but i sure as heck hope it sounds decent too.
Maybe I'm missing something, but i've been thinking a 3" driver will have ka = 6 at 10kHz, for about a 60 degree beamwidth.
The dodec has a 66 deg splay angle between faces, so I'm thinking combing might not be a big deal from 10kHz up??
And be more likely to be a big deal below 10kHz, down to maybe where c2c spacing gets relatively tight?
Swicthing gears to 'what driver.'...... a modeling question for you guys...
In trying to choose between drivers, I've been looking at each in hornresp, sealed and vented. But only modeling a single driver
I've been assuming i can just multiply volume by 12, for the dodec size. Is this ok?
But how about ports? 12 little ones, or combine into fewer and larger?
Doesn't seem like port area and length can scale linearly (like maybe volume can)?
Hope those questions made sense...
Thx, (from a guy who tries to sim as little as possible Lol)
I'd say you can multiply the volume, but for the ports there would be the option to go with fewer ports out of one shared volume. Personally I'd opt for sealed. Unless the enclosure becomes real big, chances are big they will sum favorably at say 100 Hz.
I can't say I expect it to work like you envision it for the top end. Why would you try it?
Build some more synergies and cluster them 😀.
I can't say I expect it to work like you envision it for the top end. Why would you try it?
Build some more synergies and cluster them 😀.
Attachments
Yep, i'm thinking the same way about volume and ports..
And I'd love to stay sealed if possible. Just mapping out the tradeoffs right now...
Like said earlier, the real goal behind the dodec is for measuring room acoustics. Like this man has done. Build Your Own Dodecahedron Loudspeaker
And I'd also like to learn a little more as to how the drivers couple and pattern.
But i'm truly not expecting the dodec to sound all that good....
Plus, I could not be happier with the synergy's I've built, especially the latest version. They are like giant headphones,... so tight, rich, and clear.
Paint is drying on the second one of the latest version, and I'm pulling everything else out of the test room today. It's stereo time! Lol
They are definitely my cup of tea 🙂
And I'd love to stay sealed if possible. Just mapping out the tradeoffs right now...
Like said earlier, the real goal behind the dodec is for measuring room acoustics. Like this man has done. Build Your Own Dodecahedron Loudspeaker
And I'd also like to learn a little more as to how the drivers couple and pattern.
But i'm truly not expecting the dodec to sound all that good....
Plus, I could not be happier with the synergy's I've built, especially the latest version. They are like giant headphones,... so tight, rich, and clear.
Paint is drying on the second one of the latest version, and I'm pulling everything else out of the test room today. It's stereo time! Lol
They are definitely my cup of tea 🙂
Hmm...
So, the result of a 9 point grit with 1 octave smoothing...
All of that to counter the combing that will be present.
But what can be seen is the rise at lower frequencies that I had guessed at, making a sealed solution a valid choice for this purpose?
So, the result of a 9 point grit with 1 octave smoothing...
All of that to counter the combing that will be present.
But what can be seen is the rise at lower frequencies that I had guessed at, making a sealed solution a valid choice for this purpose?
