Something like The Uluwatu but with a twist

[MaximuxKokotus]

Hi,
building my 1st set of speakers and The Uluwatu caught my eye. But I would like build it with a twist as I own a CNC machine.
1" plywood is the weapon of choice and I've been experimenting with some translam designs (while keeping the front dimensions and the volume the same). Presuming the square design is the worst one (but easiest to build) what's the right way to go? I've quickly sketched these:



Any ideas?

[JonMarsh]

You might want to take a gander at Steve Manning's posts on the TTC cabinets- he's done the conventional translam approach in the past, but I pointed him in a new direction, which I'd suggest you consider, too, as it might remove a lot of the gut ache that people go through with conventional translam builds.

So, from that viewpoint, we'd probably recommend something like you middle idea, but with some differences in the approach. Check out this thread for some details.

http://www.htguide.com/forum/showthread.php?44329-A-New-Adventure&highlight=adventure

[MaximuxKokotus]

By the "new direction" you mean the threaded rods and gaskets instead of gluing stuff together?

---

I've played a bit more in CAD and found a design with almost the same volume but the question now is how much bracing do I need? As I have to account for it in the volume. Do I even need any extra bracing?



Each speaker would consist of 39 "dual" layers and 9 "single" layers plus the full top and bottom pieces. Front would be probably also from plywood or maybe polished aluminum. Dunno yet.

[JonMarsh]

By the "new direction" you mean the threaded rods and gaskets instead of gluing stuff together?

---

I've played a bit more in CAD and found a design with almost the same volume but the question now is how much bracing do I need? As I have to account for it in the volume. Do I even need any extra bracing?

[ATTACH=CONFIG]31112[/ATTACH]

Each speaker would consist of 39 "dual" layers and 9 "single" layers plus the full top and bottom pieces. Front would be probably also from plywood or maybe polished aluminum. Dunno yet.

Yes. Steve has built translam in the past, and so have others on this forum, and others.

A major problem with the conventional translam concept is that the materials are then asked to operate in a mode in which they are not strong. Nearly EVERYONE I know that has made a plywood based translam, for example, has had issues with layers delaminating.

Consider the physics of the material properties- plywood has good bending strength, and good compression strength, but poor tensile strength in the direction of the laminations, especially when leveraged by the length of a sidewall.

MDF has good compression strength, but is mediocre in all other characteristics, including material cohesion in the precedence of leveraged force to separate a joint. This is what happens in a Translam.

I am not a physicist, but I play one on the Internet. (In real life, I’m a trained psychologist, but have just retired after working 35 years for an international semiconductor company as an engineer).

But don’t take my word for it, experiment. Use the scientific method. The TTC concept is very, very strong. And reliable.

Ask Steve. PM him.

[Steve Manning]

Ditto on what Jon said and to go a little further ...... I think you can get away with gluing the layers together, but do not glue the baffle in place. That stops the layers of the main cabinet from moving and sooner or later mother nature wins and something breaks. Take a look at Magico speakers, at least there original stuff like the Mini, V2, V3. The front and back of their cabinets are bolted together and float with respect to the main cabinet. They use o-rings as an interface between the parts for sealing. I've not read that they had any issues with stuff breaking. They have of course moved on from Baltic Birch for more exotic materials, but it can be done.

[MaximuxKokotus]

Realistically I don't even think the baffle could be glued in. The serviceability of such a speaker would be awful.
But there will be a point at which not even rods as bracing will hold (the Two Towers from DIYAudio member wesayso cracked even before they were finished). But I don't believe that will happen with such a "small" speaker.

The plan now is to draw the whole thing, cut the ribs, glue them together with some polyurethane wood glue, then seal the outside with transparent epoxy and close it with the front baffle mounted over an o-ring. It should work well I hope. Although I'm still a bit uncertain about the general shape.

[MaximuxKokotus]

Also the build specs call for "36 oz. of acousta-stuf, polyfill, or similar" what if I wanted to use bitumen like Troels does or the usual "eggcrate foam"?
How can I figure out how much of bitumen for example do I need?

[JonMarsh]

Also the build specs call for "36 oz. of acousta-stuf, polyfill, or similar" what if I wanted to use bitumen like Troels does or the usual "eggcrate foam"?
How can I figure out how much of bitumen for example do I need?

You're comparing apples and oranges.

Bitumnn is used for enclosure wall damping. That's a good thing to do, but a separate issue from internal enclosure damping.

I'd recommend staying with what Curt recommends for that- if you want to get the same results. Keep in mind that where the damping material is located is important relative to it's effect- in the main interior of the enclosure, velocity is high, and pressure low. On the wall boundaries, it's the opposite.

I'm not personally a fan of egg crate foam for lining enclosures. It's original purpose is knocking down high frequency reflections and used in isolation baffles in studios. Used a fair amount of it for that back in the 70's.

But again, use the scientific method- implement and test both, and use the solution which gives you the best measured results.

[Norm]

My experiences with laminated constructions supports the advice above that one can glue the layers so long as an element like the baffle or metal alignment/clamping rods are not also glued to the layers. On DIYAudio this thread "The Two Towers (a 25 driver Full Range line array)" is famous-in-our-niche https://www.diyaudio.com/forums/full...ine-array.html taking the laminated construction to the extreme of an floor to ceiling line array. Buried in the 681 (!) pages of posts in this project is the failure of the towers splitting some time after glue up and then necessary repairs. As several of us have tried 'wesayso' had used long lengths of all-thread rods to align and clamp the layers. My belief is his mistake was due to tight hole sizing and/or glue into interface between rod and plywood or MDF the wood and steel materials are locked together. Strong yes, but too strong when steel and wood expand at differing rates and something has to crack. Like most engineering there are a couple of ways around this.

1. Steve Manning's TTC cabinets (http://www.smjaudio.com/speakers-1#/speaker-kits) and Gauder Akustik (https://gauderakustik.com/index.php/en/) use loosely fitting rods and floating baffles. They also layer damping materials between layers. Quoting Gauder "Individual ribs made of 50 mm thick MDF material are stacked on top of each other and elaborately screwed together. In order to achieve a high level of vibration decoupling between the ribs, soft, 3 mm thin soft fiber ribs are inserted between them. This prevents the propagation of vibrations between the ribs and the music signal is not distorted." This is very elaborate, I would even characterize as 'heroic construction' or 'obsessive' take you pick. When it comes to my hi-fi I like obsessive heroic constructions.

2. I have had success gluing the layers together without metal alignment bars. In my CNC patterns I include multiple 1/4" alignment holes. During assembly I use short dowels that just span a couple of layers then switch to a different set of holes as layers are added. In this way the vertical dowels only tie together a couple of layers at a time the idea being to avoid a vertical stressor running the full height of the enclosure. This is a trade off between two requirements, 1. aligning layers during glue-up, 2. avoiding long spans of stressors trying to pull layers apart as temperatures and humidity vary. Pro tip do not underestimate the need for accurate CNCing or router work and aligning layers during glue-up. Too many hours of my life have gone to sanding and sanding and sanding and sanding after layers shifted even 1/16". I bring such pain on myself due to an obsessive love of finely finished Birch ply end grain exposed on the enclosure.

For those of us who use laminated builds this Gauder Akustik YouTube video https://youtu.be/N5D3dMeAbHE is a geek treat. Certainly emphasizes they are aiming at the top levels of the state of the art. Makes me admire SMJ Audio all the more for bringing similar results to a more attainable market segment.

[Norm]

A couple of my laminated builds.