So 9 drivers in series parallel for 8ohms and the other two at 4ohms? Both give +6dB boost at 2.83v so sounds like a win-win. Total sensitivity is probably 98dB or so?
With 11 drivers the Sd is probably close to a 12in woofer? So it’s not surprising the bass goes low as it’s all about moving lots of air to reach 60Hz.
Did you put any stuffing inside? One thing I wonder about is if there will be back wave cancellation from one driver to another. Stuffing may help that. For 12in separation I think there might be a 560Hz half wave cancellation. Since there are several paths with different lengths, the cancellation nulls will be wider spread around that. You may also have circumferential modes or spherical breathing modes that have different frequencies. So try adding some fiberglass batting.
Thanks for the photos.
With 11 drivers the Sd is probably close to a 12in woofer? So it’s not surprising the bass goes low as it’s all about moving lots of air to reach 60Hz.
Did you put any stuffing inside? One thing I wonder about is if there will be back wave cancellation from one driver to another. Stuffing may help that. For 12in separation I think there might be a 560Hz half wave cancellation. Since there are several paths with different lengths, the cancellation nulls will be wider spread around that. You may also have circumferential modes or spherical breathing modes that have different frequencies. So try adding some fiberglass batting.
Thanks for the photos.
Last edited:
XRK the graph confirms the the 560Hz cancellation that's also what i was wondering about when i asked for the c to c distance.
A yes answer to all your questions. Nice sleuthing 🙂
One piece of clarity perhaps. the approximately 12" c-2-c i said, was for the distance between any two adjacent drivers on connecting faces.
Internally, the edges of the pentagon faces are 7.5" long, so that gives a set of interior diameters per attached chart, multiple distances like you were saying.
Interestingly, neither adding 1" thick acoustic damping foam on the walls around the drivers,
or adding various levels of fiberglass stuffing,chave altered response much at all. Well, other than the fiberglass killing the low end port response.
So I dunno what to think yet about internal happenings.
What is clear, as expected, is the combing that occurs.
Below are two measurements.
Purple is a 1/3 octave FIR tuning of a single driver on-ax at about 3ft.
Green is using that tuning, but measured pointing between 3 drivers, a little further out.
Not too bad really i think, given the whacky raw response.
Back to testing.....
Last edited by a moderator:
If you are close to it, within say 6ft, pink noise clearly gains hf content as you align perpendicularly to a particular driver. 4" beaming no doubt.
Once past about 10ft away, pink sounds pretty darn consistent. Nicely so, really.
This is in a fairly large room with high ceilings, moving all around it, out in the room.
(I don't use sine sweeps, other than for measuring distortion)
Thanx for posting Mark.
I love the dodecahedron as a shape for a speaker. I have been scheming to build a stereo pr with a helper woofer in a 2-way system. I have collected more than the required 24 35mm Foster FRs (based on the original JBL). I have way more than enuff 4” “subwoofers” to do 8 in the 3rd dodo.
The video may be helpful, but somewhere i have already figured out the size and angles on the pentagons to build them. Thanx for that as well. Small enuff to 3D print, but i have already worked out in my head how to do it on my modest tablesaw.
dave
I love the dodecahedron as a shape for a speaker. I have been scheming to build a stereo pr with a helper woofer in a 2-way system. I have collected more than the required 24 35mm Foster FRs (based on the original JBL). I have way more than enuff 4” “subwoofers” to do 8 in the 3rd dodo.
The video may be helpful, but somewhere i have already figured out the size and angles on the pentagons to build them. Thanx for that as well. Small enuff to 3D print, but i have already worked out in my head how to do it on my modest tablesaw.
dave
Mark,
Your post brought back memories from one of my favorite books, the 1985 second printing of Victor Papanek's "Design For the Real World".
Maybe give some waveguides like Victor did on his dodecahedren stereo speakers a try, might fill the 500-1000Hz "scoop".
Art
Attachments
Thank you Scott, I will read the thread as soon as I can the link is much appreciated. I was really shocked by what I heard playing with my Tannoy Mini in a small-ish room, huge sound and very believable in terms of scale and image location and 3d sound very palpable better in some ways than with electrostat dipoles which I ran for over twenty years. A very high standard of SQ.
That's good, comb filtering is one of those things that's easy to measure not easy to hear, especially with music. A bit like measuring a square wave, move the mic and it no longer looks so pretty.
You're welcome. Can you describe your set up please? I experimented a bit, found it very interesting but I think my room is too small for it to work well.
That’s actually very nice frequency response - looks like the sound is going be Omni consistent anywhere in the room then.
Oh no. I should not have discovered this thread while having 22 3FE22's at hand...This looks like a nice project for a 3D printer and if successful, for a CNC.
I got to know a controversial audio writer called Jimmy Hughes. He had quite a few whacky ideas, one of which was to use reflected sound by pointing his speakers towards the wall behind. I went to his house and heard his set up. Sounded really good.
Sorry I have been distracted with life issues and not watching here . What happened to Scott Joplin?
The ones i played with were 20-sided, an icosahedron.
I ran into one recently, went down and counted the drivers i have for my planned dodo, and i have 50 so...
dave
He got booted for continuous trolling and several subsequent (rather transparent) attempts at sock-puppeting. A shame, given that when he wasn't deliberately trying to stir up trouble, he could come out with something useful.