[noah katz]

Cool stuff you got going there.

Re braces, isn't it pretty straightforward?

Seems like the driver/PR rods take care of the front/back panels, and cross-rods for the sides, and maybe one central one for the smaller top/bottom panels.

Or you could truly engineer it and use FEA or Roark's plate vibration formulas to calculate the rod spacing required for your desired minimum panel resonance freq.

 

[esappy]

You can get the required panel stiffness with dowels between opposite panels.

I was going to suggest the same thing, but using hardwood dowels; you can use long deck or lag screws that will each have hundreds of lb pullout capacity.

That removes the cosmetic concern with all-thread and nuts.

Not sure I get the concern with sufficient torque on the driver fasteners.

No real concern with sufficient torque per se. That's just 20+ years of being a Machinist Mate in the Navy, plus another 6 years as a project manager for the Navy. Any time you see any kind of fastener, you ALWAYS assume there is a torque spec that goes with it. That was more having fun with Chris since he is an engineer.

I have thought about the dowel thing as well. That would give it a sort of 'stonehenge' kind of look for interior bracing, which I like, and I like the idea of the pulling force that comes with it as long as you line up the subs bolts holes with the dowels. Question though Noah. would you prefer long deck or lag screws over using threaded inserts and machine screws? Do you think the threaded inserts would strip out over time with the amount of subwoofin' that will be going on in this case? Just curious.

 

[esappy]

In all seriousness, I like a bit of an industrial look. Maybe my time spent working for the navy's submarines, but I appreciate a proper bolting job. This thing is going to look awesome when it's done, but keep in mind I'm not the prettiest thing that graced this earth either so maybe birds of a feather....

I was thinking more of a steampunk kind of theme, but I agree with you about liking the industrial look.

I too worked on submarines! That's awesome. Small world for sure.

 

[esappy]

I'm also an ME; I spent 30 yr at Lockheed doing space instrument optomechanical and structural design, and vibration analysis and testing.

OK, well that explains it! If you have time in the near future, I would love to pick your brain about bracing and vibration reduction. Suddenly I'm feeling like the dumb guy in the room since I was only an engineering technician.

 

[Chris Popovich]

ha! Never feel like the dumb guy when you're with a bunch of engineers, I am one and love 'em, but half of them wear velcro shoes, the outrageousness of which is only exceeded by the lack of social graces. It's ok though, there's room for all types. I've made a bit of a specialty of getting the tech types and the hard engineers to play together better. Different languages really, but I definitely want both types on my teams. Head in the clouds absent minded professor capability with some wrench turning artistry and we can do amazing things. Best combo ever.

 

[esappy]

ha! Never feel like the dumb guy when you're with a bunch of engineers, I am one and love 'em, but half of them wear velcro shoes, the outrageousness of which is only exceeded by the lack of social graces. It's ok though, there's room for all types. I've made a bit of a specialty of getting the tech types and the hard engineers to play together better. Different languages really, but I definitely want both types on my teams. Head in the clouds absent minded professor capability with some wrench turning artistry and we can do amazing things. Best combo ever.

Agreed!

 

[JohnDean]

More speaker building and cool pictures. Less talk about cool jobs that I drank too much in college to get a degree in.

 

[noah katz]

Question though Noah. would you prefer long deck or lag screws over using threaded inserts and machine screws? Do you think the threaded inserts would strip out over time with the amount of subwoofin' that will be going on in this case? Just curious.

Like I said I like buttonhead screws, so if it was me I'd start with long ones (say, 3") threaded directly into hardwood dowels.

I doubt it would be necessary (max load per fastener isn't really that high), but as a relatively easy insurance policy that could be beefed up by wetting the threads with superglue and then re-tapping.

If the threads did feel, they could be drilled out and internally threaded metal rods coul be glued in.

OK, well that explains it! If you have time in the near future, I would love to pick your brain about bracing and vibration reduction. Suddenly I'm feeling like the dumb guy in the room since I was only an engineering technician.

Sure, though all the basic qualitatively favorable design concepts are well known around here; after that it comes down to doing actual calculations.

You can get a free personal use Fusion 360 license and have at it with FEA, as well as CAD.

 

[Chris Popovich]

Clamp design done, getting quotes and a sample being made next. Already ordered the hardware. This will allow easy attachment from inside the box, the double holes are for the two (slightly) different bolt circle diameters between the LaVoce and the PR's. It would be easy to add another bolt circle diameter or two should anyone want to follow suit. Just need 8 for my test box/proof of concept, after that I'll be buying about 50 of them haha.

 

[Chris Popovich]

Alright. Update. The clamps are done and work well.

Guilding the lilly; 21" subwoofer & PR clamping...

Hey AVS brethren, While in process with my dual 21" dual PR subwoofer design, one of the issues I had was how to attach the drivers to the box. It's true that you can just spend a couple bucks for some big screws and that works pretty well most of the time, but I've had the wood come apart...

www.avsforum.com

Update to the main project is that I've acquired the rest of the active drivers and PR's. All speaker "stuffs" is done and ready. Box clamping mechanisms are done and ready -at least for the first two subs. Will have to order more for second two but that's nothing to sweat. I have plain hardware on hand, and some fancy stuff.

Box volume will be about 12 cuft. Currently working through design iterations trying to put a little prettiness on this big girl, then brace analysis will begin.

Fitment analysis is done -couple pics attached where, while considering a range of front baffle widths, I wanted to ensure that no interference would exist with the actives and PR's. Resultantly, models were made.

It is tight, but by locating the PR's dual-opposed between the active drivers, we can use a config that would not be possible were all the drivers to have the same altitude.

All in all it's slow but work has definitely gone on and I'm pleased with the state of the project -this was never intended to be a quick build.

Chris

 

[Chris Popovich]

Very early concept of final enclosure. Compromises will be made in cost and footprint to have some curves but this pig needs lipstick.

 

[Chris Popovich]

It's a trapezoid, but only just so.

 

[Chris Popovich]

Progress continues. Below is an excerpt of the reporting being done as this project moves forward (it's very well organized even though I've been struggling with the details of how to make the box look), and an example of how optimization of box curves might be done in a way that is useful in the context of not reducing wall thickness too much and preserving at least the inner layer dados. The note about the front side on the bottom is incorrect, it's actually on the right side. Drivers will be recessed, so structurally speaking the PR's will be bolted through double wall and the LaVoce's through single wall. We'll gain some of the volume back by machining out parts of the innermost wall when/where/if appropriate.

Next step is to decide actual driver frame cutout (for recessing), pick a small visually pleasing radius for that edge, then attempt to minimize baffle width for chosen driver cutout + wiggle room + recess radius (x2) + box corner radius (x2). This will give us a minimum baffle width range to play with. I'll make sure that doesn't interfere with the clamp system on the inside diameter and at that point we've fixed those parts of the design.

Anyone care to hazard a guess which r I'm choosing?

Oh, and for the very front panel (really just visual) it'll most likely be located with dowel pins, so it's not just a piece of wood face glued to another one.

 

[Chris Popovich]

Slow and steady. Just bought some Revel F226BE's, and been having some health issues so the sub project got put on back burner for a while. We're back up and pushing forward.

 

[Russdawg1]

2.9 or 3.3 on the radius?

 

[Chris Popovich]

3.3

 

[(The Other) Colin Miller]

I like where this is headed!

 

[Chris Popovich]

One step back, several steps forward. Will volume-reduce bracing once all is finalized location wise.

You can see the (slight) trapezoidal shape and those sweet curves when looking top down.

 

[Trimlock]

I’m sure you get asked a lot, what is the 3d modeling software you are using?


Sent from my iPhone using Tapatalk

 

[Chris Popovich]

Not sure. I'm not a CAD guy, so I went snooping around on UPWORK and put out a proposal for the old quad-12 JBL project back in the day, found a cool dude in Montana that does my CAD work for a reasonable fee. Smart guy, I've set it up such that project progress is being done both informally via email exchange of information as well as in sequenced reports. This is probably the third project I've used him for, and by far the biggest in that many aspects of this design were unknown when we started. I don't mean box volume, I mean how we'd make that volume and what footprint/shape it'd be.

This is primarily due to wanting to make this "pretty". In the past it was always simple; make the box, double layer the front/recess the driver, add a couple dados and build the damned thing. Duratex and you're done. This is different where I'm pushing limits on how much curvature I can get, and fit the the drivers with no interference, and have my clamp mechanism not hit the walls. The audio part of the design (drivers, PR's, mass/tuning and box volume) were the easiest part of the design. It is due to the unknows that I had to take an analytical approach and actually see what could be done curvature wise -and at what expense. As such, I have run this project like an engineering project, weekly check ins, reports, milestones, etc, but fun since I'm the one in charge and not the corporate pressure cooker wonks I'm more familiar with. Photos of hand sketches, MS Paint being abused to mark up nice drawings, it's been fun running it like this, plus I have the "thought process" well documented for the analytical approach used for stuff that bends limits a bit. For example, have front facing drivers and want some side firing ones too? How close can you make them? Easy to think about, and if we're just fooling around it's easy to eyeball them, but really how would you figure it out if you had to make it VERY close?

Well, like this (partly):

Of course I added complexity by staggering vertical location, and actually making a model of the PR, but you get the point.

I could have built this sucker months ago, but I'm having fun seeing what flexing some engineering looks like when the final project is completed vs. slapping a box together. Will it be worth the effort and expense compared to a flat pack built in an evening? I guess we'll find out.

Bracing is about 60% done, little kiss here and there and volume reduce, then we'll do detail work; label plate, terminals, dowel pins for locating purposes (maybe), feet, etc.