Post 193 will get you the dimensions based on volume and post 191 shows the steps. It is different than the 1 sheet PDF above as the baffle is a square with edges of length S and all other edges are 0.866 x S. You have 1 square, 4 triangles (each with edges S, and qnty 2 x 0.866 x S), and Qnty 8 rhomboids of edge 0.866 x S. Really a 13 sided speaker and not a true dodecahedron.
Last edited:
xrk, I really like your foamcore creations. It's such a great material for quickly putting something together. I thought you might be interested in this thing if you don't already know about it. It's a rabbet cutter, to easily make nice corners. I used one for years for industrial design mockups. Here is a link to one retailer that sells them:
Foamboard Rabbet Cutter - BLICK art materials
Foamboard Rabbet Cutter - BLICK art materials
That looks neat and a good tool to have. Is a rabbet a 90 deg notch on the edge that lets the corners join with the paper from one board covering the exposed foam? It would be neat to have a dovetail joint press.
That looks neat and a good tool to have. Is a rabbet a 90 deg notch on the edge that lets the corners join with the paper from one board covering the exposed foam? It would be neat to have a dovetail joint press.
Yes, after you join two pieces at right angles with a rabbet, all the foam is covered and you just see the paper. Now as for dovetails, you're on your own...
Hi,
Its pretty clear to a 1st approximation the 8 rhomboids
main nodal resonance will coincide and affect the output,
and their effect depends on the phase sum measured
at an angle and a distance, but they are clearly there
in the responses, and the impulses, for which they
clearly carry on for a long time, indicating high Q.
The speaker will have no resolution in that area,
particularly if further modes are harmonically
related, and they inevitably highly likely will be.
Its a nice shape for outside diffraction (but still needs
BSC) and and a nice shape to complicate internal
standing waves a lot, due to the internal angles.
However there is nothing mystical about it, and
it succumbs to its limitations like any other
speaker enclosure shape, that need addressing.
Here I'd suggest some variability in the 8 rhomboids,
say centre damping 4 of them, reducing the symmetry
will bring edge effects more to the fore, and so a lossy
edge glue like rubber cement could probably help.
IMO destroying the mechanical symmetry would help a lot.
Various flat stick on rubber feet * placed off centre on each
foamboard panel would introduce an element of chaos to
the way the whole thing interacts, and smooth response.
rgds, sreten.
* might sound crude, but that is exactly what Mission
NXT speakers used, albiet small and computer placed.
Easy to improvise with some damping sheeting.
Its pretty clear to a 1st approximation the 8 rhomboids
main nodal resonance will coincide and affect the output,
and their effect depends on the phase sum measured
at an angle and a distance, but they are clearly there
in the responses, and the impulses, for which they
clearly carry on for a long time, indicating high Q.
The speaker will have no resolution in that area,
particularly if further modes are harmonically
related, and they inevitably highly likely will be.
Its a nice shape for outside diffraction (but still needs
BSC) and and a nice shape to complicate internal
standing waves a lot, due to the internal angles.
However there is nothing mystical about it, and
it succumbs to its limitations like any other
speaker enclosure shape, that need addressing.
Here I'd suggest some variability in the 8 rhomboids,
say centre damping 4 of them, reducing the symmetry
will bring edge effects more to the fore, and so a lossy
edge glue like rubber cement could probably help.
IMO destroying the mechanical symmetry would help a lot.
Various flat stick on rubber feet * placed off centre on each
foamboard panel would introduce an element of chaos to
the way the whole thing interacts, and smooth response.
rgds, sreten.
* might sound crude, but that is exactly what Mission
NXT speakers used, albiet small and computer placed.
Easy to improvise with some damping sheeting.
Last edited:
That looks neat and a good tool to have. Is a rabbet a 90 deg notch on the edge that lets the corners join with the paper from one board covering the exposed foam? It would be neat to have a dovetail joint press.
If you click on the "Photos" tab on the referenced product page, you will find a photo of it in use.
For making folding joints, you need a cutter that takes a v-groove out of the material leaving the backing paper on one side intact. Something like this:
Logan WC-2001 FoamWerks V-Groove Cutter - Logan Graphic Products
That V-groove cutter is neat; I hadn't seen that before. Alternatively, you can just make a regular straight cut through the inside paper and the foam, leaving the outside paper, then run along it with a dull point such as a retracted ballpoint pen to make the V. We used wooden dowels with turned ends.
Petty, squabbling posts removed, the guilty parties are warned that continued posts on this sort will result in points or bin time being awarded.
I would like to point out that the truncated rhombic dodecahedron does indeed have problems in some applications. Here is an example. I apologize that I do not have access to the data that resulted in the decision of the parties involved.
In the late 90s I was in communication with the leadership of the Hardees chain of restaurants. They had experienced problems with their drive-through intercom systems and were looking for ways to improve them.
As a result of our conversations they agreed to give the square-truncated rhombic dodecahedron enclosure a test. They therefore sent us one of their exterior speakers (which their system at that time also used as a microphone) and the associated tall parallelepiped post it is ensconced in. Knowing that this system can be exposed to weather, we produced a stainless steel enclosure and sent it to them for testing.
Unfortunately, their testing resulted in their declining to deploy this design in their systems. The reason for this that was relayed to us was as follows. When they connected the RD-enclosed driver in a franchise to test its usefulness, they found (since it was used both as a microphone and a speaker in their intercom system) that it was too sensitive -- (!!) it picked up so much of what was going on in the vicinity that it was counterproductive for them to use it. They needed to receive more directionally in order to focus on the driver placing an order.
In the late 90s I was in communication with the leadership of the Hardees chain of restaurants. They had experienced problems with their drive-through intercom systems and were looking for ways to improve them.
As a result of our conversations they agreed to give the square-truncated rhombic dodecahedron enclosure a test. They therefore sent us one of their exterior speakers (which their system at that time also used as a microphone) and the associated tall parallelepiped post it is ensconced in. Knowing that this system can be exposed to weather, we produced a stainless steel enclosure and sent it to them for testing.
Unfortunately, their testing resulted in their declining to deploy this design in their systems. The reason for this that was relayed to us was as follows. When they connected the RD-enclosed driver in a franchise to test its usefulness, they found (since it was used both as a microphone and a speaker in their intercom system) that it was too sensitive -- (!!) it picked up so much of what was going on in the vicinity that it was counterproductive for them to use it. They needed to receive more directionally in order to focus on the driver placing an order.
I finally got a chance to listen to some of my standard test tracks with the Hypercubes in stereo. They sound really good despite the measured peak at 2.2kHz. Set on a stand and 4 ft from a rear wall, the imaging is extremely good, with a larger than expected soundstage. Presumably this may have to do with some of the enclosure wall emission that gives more ambiance. It in some ways sounds almost like a 3d DSP sound processor was turned on - but in a good way. One really interesting thing was that the speakers sound a whole lot better when aimed at 45 deg up vs perfectly horizontal. This really changed the character and soundstage in a very positive way. I also noticed something new: on some recordings, the level of background ambient sounds in the recording are much more apparent. Like the sound of audience whispers, shuffling of feet, squeaky chairs, etc are more perceptible. In a way I felt like the listening experience was closer to live with the performers and the other audience members sitting next to me. I have to say that this was unexpected as I have heard this recording hundreds of times before and with the same driver but in different enclosures, and in the same room with the same amp, with the same source material and CD player. So I know it is really just due to the speaker. Given this, I have to say that there is something very special when you listen to them in stereo and appropriately mounted on a stand and aimed up at 45deg (a natural placement for them to be stable). I think that the measurements so support the fact that there is a lot of secondary broadband emission, and this, really makes this speaker sound great. I highly recommend that anyone having 3 hrs to try this out on Foamcore and a low cost full range driver. The TC9FD is great but a Faital Pro 4FE35 or 3FE25 would also be excellent choices. Certainly with more expensive full range drivers like a FF85WK or FF125WK, I think that these would work very well too.
The extra audibility of the recorded background material would seem to confirm Tesserax's experience with Hardee's chain in saying that as a microphone, they pick up too much ambient sounds - a speaker and mic are reciprocals.
This is a good thing for speakers (or an ambient sound omni mic).
The extra audibility of the recorded background material would seem to confirm Tesserax's experience with Hardee's chain in saying that as a microphone, they pick up too much ambient sounds - a speaker and mic are reciprocals.
This is a good thing for speakers (or an ambient sound omni mic).
Last edited:
I'm glad your experience is matching mine. I was really impressed by the purity and "non-directionality" of sound with one enclosure but when I finally listened to a properly set up stereo pair I was absolutely floored. I have mine slightly tilted back and aimed right at me (I'm looking right at the tip of the phase plugs) and with a little dsp (less correction strength than I was using with my 8" 2-ways) the resulting sound is absolutely spectacular. The soundstage is so big it's ridiculous and I just can't get over the bass quality and extension. I can't express how much I love listening to music with these little things. I thought about trying the upward facing position but I just don't want to touch anything right now.
When I get home tonight, I can show you uncorrected and (gently) corrected measurements made from my listening spot which is 5' from the speakers.
I use DRC 3.2.1 by Dennis Sbraigon which analyzes an impulse response and generates a correction filter in the form of a .wav file which gets convolved with the music signal in real time (using Foobar music player). Improvements are clear in both the frequency and time domain. I use the "minimal" preset and edited it so I would have flat extension down to ~64hz.
That would explain why your earlier freq response measurements were so flat. Can you get a measurement at same position with RAW no DSP applied?
I think maybe I wasn't clear....the measurements that I have provided had no dsp applied to them. My polar plots were shown with 12db smoothing unlike my waterfall plots which is why the waterfalls are a bit rougher.
Thanks for clarifying. They really are flat. Looking forward to measurements at 1m or listening position.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.