I can make add another layer of MDF, which should make it extremely stiff as well as strong. Well, putting in the necessary washers to support the clamping load may be slightly tricky.I would be tempted to make the lower of the two mouth rings from aluminium. I have seen thicker MDF rings of that size distort. Also a lot of open end grain MDF will absorb moisture.
The compression from bolting the basket legs to MDF will make it hard to maintain exact symmetry. Especially after bolting them on and off a few times and with the weight stress.
Cool project btw...what are your thoughts on modifying (butchering) a suitable existing driver for experimentation?
That much aluminum is very expensive... all these parts will add up to over $1000 already.
At least $250 CAD for a 24"x24" plate of 1/2" aluminum.Yeah.. It could get spendy quick. I was thinking 10mm 4000 series alu for the base mounting ring only. Water cutting a simple dounut shouldn't be to bad. Keep MDF on top layer so you can shape it.
But it does all add up ;P
I can't see that anything will excite modes of vibration in the cone aside from the force on the voice coil itself. The compression driver is clamped into a fairly solid steel plug.
I was thinking of the actual sound pressure from the compression driver inciting vibrations in the cone of the WG itself, which then re-radiates at a different frequency, thereby increasing harmonic distortion. This is what I saw with the tractrix horn until I damped the horn walls with constrained layer damping using latex caulking sandwich. Maybe no concern if your CD is highpassed high above the mechanical vibrational modes of the main WG cone. I was using a full range driver down to 300Hz.
Interesting. Then the same phenomenon should be possible for a woofer to incite resonance in a mid or a tweeter in a 3- or 2-way standard speaker, no?I was thinking of the actual sound pressure from the compression driver inciting vibrations in the cone of the WG itself, which then re-radiates at a different frequency, thereby increasing harmonic distortion. This is what I saw with the tractrix horn until I damped the horn walls with constrained layer damping using latex caulking sandwich. Maybe no concern if your CD is highpassed high above the mechanical vibrational modes of the main WG cone. I was using a full range driver down to 300Hz.
It should be simpler to go with a single layer of spun aluminum and epoxy on some strips of this stuff on the back. The weight is basically negligible but the increased rigidity should make the cone insanely stiff.I can't see that anything will excite modes of vibration in the cone aside from the force on the voice coil itself. The compression driver is clamped into a fairly solid steel plug.
The cone should be made out of spun aluminum, I suspect. As for increasing the modal frequencies, the two options open to me are making two cones out of a thinner material and sandwiching a structural foam core in between, or, adding stiffening ribs to the backside, a la the Kef Uni-Q woofer.
Carbon Fiber Foam Core
0.25" Carbon Fiber Balsa Core Texture 06" x 18"
Interesting. Then the same phenomenon should be possible for a woofer to incite resonance in a mid or a tweeter in a 3- or 2-way standard speaker, no?
Just depends on natural resonant frequency being in band of sound output from one driver to the other. Probably not a problem for a woofer to excite a tweeter because its fs is so much higher. With the tractrix or in this case, a coax it is more of a concern because the source is right inside and adjacent to the cone itself. If the natural resonant freq of the WG cone is much lower than the CD range there won't be a problem. If however, there is a resonant freq say at 1500z in the WG and you have a CD that covers 1000Hz and higher then there may be some coupling.
Might be an issue. Seeing as how they cross over to each other, the overlap is unavoidable.Just depends on natural resonant frequency being in band of sound output from one driver to the other. Probably not a problem for a woofer to excite a tweeter because its fs is so much higher. With the tractrix or in this case, a coax it is more of a concern because the source is right inside and adjacent to the cone itself. If the natural resonant freq of the WG cone is much lower than the CD range there won't be a problem. If however, there is a resonant freq say at 1500z in the WG and you have a CD that covers 1000Hz and higher then there may be some coupling.
It's a great cone material but I don't know how to form it.Wow... $250.. Tempting to take a grinder to a second hand cast frame woofer. Anywho..
Was it Vandersteen who uses skinned balsa cones for his MIDs?
Thanks, I'll check them out.Po Yun/Precision Econowind?
In order to make the cones easier to fabricate, I'll be changing the cone profile from OS to conical. After the first few inches, it looks like there is very little difference, and the benefit is that I will be able to take a sheet of aluminum and form the cone by rolling a cutout, like so:
Flat Cone Template Calculator
This way, I am also not limited by the alloy that can be used, as would be the case for metal spinning, nor am I limited by a minimum thickness. It should also be easier for me to fit up the cone to the voice coil former for a seamless joint. Plus, I save a whole bunch of money by not having to make a mandrel or pay someone to spin some sheet aluminum for me (and possibly screw up in the process).
Since I can (and should) use very thin sheet with the rolled-cone construction, it makes sense to look at going with a sandwich construction again. Now, as for core materials, I don't want to use anything rigid as that means I either have to use heat and/or pressure and/or water to get the material to conform to the shape of the cone. Luckily, there are flexible core materials out there, e.g. Nomex honeycomb sheet. I can cut it out the same way I cut out the first cone, put on some epoxy, and wrap it around. The second cone can be made after I measure the relevant ODs of the wrapped honeycomb layer.
Since the weight of the core is fairly negligible, I can make it arbitrarily stiff simply by increasing the thickness of the core. It would be a dream come true to get the first breakup mode at least 2 octaves above the LP frequency, but unfortunately I don't have the ability to simulate that with FEA.
Last edited:
I'll be changing the cone profile from OS to conical. After the first few inches, it looks like there is very little difference ...
Its the first few inches that are the most important however.
Yes, which is good because that part, the combination "center pole/waveguide throat", will be handled by a CNC lathe.Its the first few inches that are the most important however.
I've attached a worksheet that helps me get an idea of what I can expect as far as sensitivity goes. The actual sensitivity will be somewhat less, due to mass of adhesive and core, and virtual mass from the spider and surround. Hopefully I can get the flux density up somehow. Measured sensitivity will also be closer to half-space rather than full-space, since the size of the waveguide itself dictates that a very large baffle is used. Adding a little bit more baffle should allow at least a 3dB boost over the full-space figure, but since active EQ has to be used, a little bit of efficiency loss at the lowest part of the passband isn't a big deal.
Attachments
Last edited:
Yes, which is good because that part, the combination "center pole/waveguide throat", will be handled by a CNC lathe.
That works. Now to solve the other hundreds of problems.
Void-free BB plywood is also an excellent alternative to the MDF. Stronger, less dense, and more stable vs changes in humidity.Wow... $250.. Tempting to take a grinder to a second hand cast frame woofer. Anywho..
Was it Vandersteen who uses skinned balsa cones for his MIDs?
Yes, if I get them built I would have to have someone else measure them, since I'm terrible at it. We'll see...If you get them built, I'd be happy to play with a pair for you. You know, as a favor. Ambitious project indeed.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
- Home
- Loudspeakers
- Multi-Way
- 18" 90-deg OS WG coaxial (design)