Unconventional Speaker Design - thread carried over from multiway forum

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Update:

Started glueing my blocks together. Using some elastic instead of string to hold it together while it dries. I'm pretty sure it'll be strong enough, especially once I get layers of glue, and nails from the top/bottom. It's also a lot more even now, got my technique down well.

An externally hosted image should be here but it was not working when we last tested it.


You can see my prototype in the background - my new wood has a much nicer colour!
 
The issue that you had with the circle breaking when squeezed from 2 sides won't be an issue when you start stacking the rings on top of each other. The extra layer on top will give you strenght in a differant plane, thus giving it strength in more dimensions. If you add the front baffle, even more strength.

The glue will work fine, you just need to compress it a bit till it dries, a day should be more than enough if there is not too much moisture in the air.
When glueing the rings to each other see if you can't get hold of some clamps (maybe 4) to really apply pressure. Just put a piece of wood between your workpiece and the clamps so that the pressure is distributed evenly.

Also try applying the glue with an artist paint brush or small scrap of wood, cover the whole surface thinly and evenly. I always try and avoid huge blobs, the just get forced out on the sides when compressing the pieces together.

The vent is no problem, you can just use the info what they give on the madisound website, with some fine tuning a bit later on.
 
PVA glue is fine if you know how to do it. ( I prefer Titebond) Work it into the grain on both surfaces, full coverage, clamp. It is stronger than the wood. Weights, tape, calmps, string. Lots of options. Brad nailers are handy as your blocks are likely to slide around.
 
I've only done enough woodworking to be dangerous, but any glue is weaker if it has to fill a significant gap. I'd worry if these pieces aren't cut to high accuracy so that all joints aren't perfectly flush. I'd even want to sand down the top and bottom of each ring (lightly glue sandpaper on a flat table or large/thick glass plate and slide a glued and dried ring over it to sand down and flatten out any high spots) before putting them together.

Here's an interesting article on wood glues:
http://www.oldbrownglue.com/pdf/HowStrongisYourGlue_FWW.pdf
 
Caveat: I'm also in the "done just enough woodworking to be dangerous" category. However, that said, it looks to me like you are gluing endgrain to endgrain. This notoriously gives a very poor bond. Splines can be used to get more of the bondable side grain into the glue surface. If they are flat (horizontal) splines, the glue surfaces should be all side grain and should be the next best thing to a finger joint, or box joint. Vertical splines can be used too, but that gets a bit dodgy at an angle like that, according to *some* (me for instance, still deep into my learning curve...). Just had a very experienced woodworker build me a cab with angled vertical splines like that. He thinks it's fine.

He's also not at all concerned about wood movement on a cab this size- I've met other experienced woodworkers who are deeply, deeply concerned about wood movement. They convinced me to be deeply concerned too. Movement along the grain is very slight, but across the grain it can add up. The way yours is laid out, it looks like the grain is horizontal. I see this as good news in that your "circle" won't be getting bigger and smaller in any significant way, so that should be stable longterm. The expansion should be almost all in the vertical direction, across the grain. So your cab will get taller and shorter by some small fraction of an inch as the seasons change and as humidity changes.
Keep this in mind as you go to mount your baffle. Plywood, which is almost always used, since it's stable and doesn't change dimension much and resists cupping, will not be moving along with your solid wood. One theory- it'll keep the solid wood from moving. Another theory- something's gotta give! And since the plywood is only on one side, the back side will be expanding and contracting vertically while the front side either can't or busts free of the plywood.

Now to the part where the glue is usually stronger than the wood- so the wood breaks, when something gives. Another alternative for the old school fanatic is hide glue- not the TitebondTM liquid hide glue, which will fall apart with any significant humidity, but real cook it up in the hot pot gelatin based hide glue. My violinmaker says the best stuff is generic gelatin from the supermarket. Now we have glue that breaks before the wood does, and is flexible long before it breaks.

Right now I'm having cabinets built with solid wood baffles. This is not usually recommended, as they are heir to all the woes of the unlaminated flesh, but they will move along with the rest of the solid wood cabinets, provided the grain is oriented the right way. With quartersawn wood or wood joined cutting board style, cupping is minimized.

Once again, I'm not an experienced cabinetmaker, I'm just a guy who spends way too much time working with cabinetmakers, specifically to have them build me relatively stable speaker cabinets out of solid wood. Mine are instrument cabinets, so I welcome the harmonic contribution from the solid wood. I've talked to a lot of very experienced craftsmen about all this, and as I mentioned they don't all agree completely and are often passionate in their disagreements.

I don't think there's much controversy over gluing endgrain to endgrain, though. That's notoriously weak. Probably why they are busting apart so easily.
 
Thanks for the link on wood glues! That's very useful for me. Aslo very nice to have something well out of the realm of anecdote and opinion. I've never used this "Old Brown Glue"- hopefully it's a lot more resistant to humidity than that Titebond stuff. I had some Titebond liquid hide glue joints fail in a humid indoor environment, under no pressure- they just fell apart.

The one thing this article does not address is relative wood movement. I'm going to see if I can find anything on wood glues and relative wood movement.
 
Lots of good stuff on hide glue at player-care.com, a site dedicated to restoring old player pianos. Apparently there's an issue of "joint creep", that may be worse with synthetic glues. I'm going to email some of these glue makers and find out the relevance to my concern- speaker cabinet joinery.
 
I've only done enough woodworking to be dangerous, but any glue is weaker if it has to fill a significant gap. I'd worry if these pieces aren't cut to high accuracy so that all joints aren't perfectly flush. I'd even want to sand down the top and bottom of each ring (lightly glue sandpaper on a flat table or large/thick glass plate and slide a glued and dried ring over it to sand down and flatten out any high spots) before putting them together.

There's a number of reasons why my prototype was a bit weak - this is one of them: not all the pieces were flush, due to some bad cutting technique. It's totally fixed for the real thing though. And I will definitely sand each ring down before glueing.

Caveat: I'm also in the "done just enough woodworking to be dangerous" category.
//
Probably why they are busting apart so easily.

I appreciate your input. Not much I can do about the grain. But each piece will be glued to its left, right, top and bottom, so I don't think it'll be a huge problem in the end.

Also, in terms of vertical wood warping, I don't think this'll be a huge issue. Firstly, the wood is second hand and quite old, so hopefully has done most of its warping (correct me if I'm wrong here), and wasn't warped badly at all before thicknessing/cutting into planks.

Secondly, I'm planning on making each ring from a U into a D shape, using the leftover wood I've got. Then I'll stack each D on top, with the flat part of the D forming the baffle. This'll allow the speaker to move vertically without affecting the strength of the box as a whole at all. If this is a bad plan - let me know ASAP, I gotta start cutting the rest of the wood tomorrow night.
D's like so, apologies for the world's worst blueprint:
An externally hosted image should be here but it was not working when we last tested it.


Update pic:
An externally hosted image should be here but it was not working when we last tested it.

The flash really brought out the ugly glue stains. I'll sand them off. Another question - should I leave the rings in this polygonal shape, or sand it into a perfect semicircle? Aesthetically I think it'd look nicer as is, but this might affect the sound.
edit: these aren't glued together! Don't worry!
 
Last edited:
I leave the rings in this polygonal shape, or sand it into a perfect semicircle?

You can sand it down to a perfect semi circle if you have about a gazillion hours spare, but I wouldn't do it, the sound won't be affected in this application. When you get to the baffle you can round the edges, this will have a far greater effect than the shape of the blocks inside/outside
 
Speakers came today. Look awesome. Gotta get an amp for them.. suggestions, for something cheapish? I've got a decent one lying around, but its quite big. I could always just put it under the desk.

I'm going to cut and glue my D's tomorrow, unless someone can give me a reason not to. Also, I think just glue, not screws. Screws will just be from the top and bottom.
 
Update:
An externally hosted image should be here but it was not working when we last tested it.


I didn't cut the D's after all, because the original thicknessing on the wood was not totally accurate, so some planks shortened from 44mm to like 40mm at points. Now they're all about 38mm - I'll glue 5 planks together, then trim off the excess.

Unfortunately one of the wood planks had cracked a bit. Not sure if that was there originally or developed in the shitty storage room I was using. I've used that plank entirely for the top panel, as I figured it would matter least there. Unfortunately I don't have enough wood to totally discard that plank.

I would appreciate any advice as to how to do this vent. I honestly have no idea how to do the calculations (the volume of my box is slightly different to the plan, and the driver itself is slightly different too). Along with calculations I'm not sure how to fit the vent into this oddly shaped box. I'm going to upload a blueprint as to what all the measurements are ASAP.

edit:Blueprint (not to scale, but measurements are good - within a few mm):
An externally hosted image should be here but it was not working when we last tested it.

Volume is around 2.7L.
Speaker is http://www.fountek.net/products.asp?id=52.

I don't mean to palm this off onto you guys, but the reality is I have very little idea what I'm doing, and only a week and a half left to do it (while struggling to pass my other subjects). I'll try to find out the details on my own, but I'd really appreciate any help with this.
I'm planning to stick with the 10mm horizontal vent described in the original diagram, unless there's a good reason not to. This does get a little difficult when it comes to the sides of the box. My wood is 18mm thick, and this'll mean it'll have to fit alongside TWO of the semicircles, which alternate, so it'll have to be really weirdly cut. Does the vent have to be flush with the sides? I assume so.
 
Last edited:
With your baffle (flat) face planks oriented vertically, like in your photo, they won't expand (not warp) vertically. The wood fibers don't get any longer. But the rest of your cab will expand vertically. So I would recommend running the baffle planks horizontally as in your "world's worst" plan diagram.

I don't know how this vent is constructed, but you may be able to tune it by changing the length. What was the original volume supposed to be? And it was designed for this specific speaker originally?
 
Unfortunately it's all cut and glued already. I do wish I'd gone with them horizontally, but the original thicknessing of the wood was not accurate on the wood I have left - dipped in by 2-3mm here and there, so I didn't really have a choice. Also, now I'm screwing into the endgrain. Ugh.

If the warping is going to be a HUGE issue, I could slice the planks horizontally and glue them back together? I'd rather not though, I don't have much time left.

The original plan is here, using an FR88EX. I have an FR89EX, which is extremely similar. The volume of both the diagram and mine is around 2.7L. I was planning to use the "alternate vent 10mm high, width of box" described
An externally hosted image should be here but it was not working when we last tested it.
 
Will a circular tube be alright as an alternate vent? To get the same area it'd be a 36mm circle cut out. Can I then just use plastic (76mm long, as in the diagram) for the tube? Or do I need to make this vent out of wood? And where does it go, can I put it roughly at the bottom, or does the position of the vent matter a lot?
 
I don't think the location of the vent matters much. A tubular vent with the same cross-sectional area a similar length should work just fine. However the handiest thing to use will be some PVC or ABS pipe, and it only comes in certain sizes. However if your numbers are right, 1 1/2 inch ID (inside diameter) pipe is nearly perfect. (36 mm=1.524") There is an "end correction" to the equations for determining duct lengths that is different for unflanged and flanged ducts- they're all "flanged" on the outside of the box, assuming they are flush, but the pipe just sticking out into the internal space is a different situation than the triangle wooden duct illustrated, so the duct length may vary. However, tuning is easily accomplished by trial and error using different lengths of PVC pipe. This may well account for the differences between the specified driver and the one you are using as well. Consider going with a PVC duct so you can tune it by trial and error. When you get the flattest response, or the one you prefer, you're done tuning.

As for the wood movement issue with the baffle- lots more stable to have the wood all moving together, i.e. grain oriented the same way. Someone told me recently that the ancient Egyptians used to use swelling wood to split stones. It can easily break itself, depending on the environmental humidity variations. Finishing both the inner and outer services will help the wood not to take on much moisture. It'll last long enough for you to get through your class. It's just how many more seasons it lasts that's in question.

Good luck!
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.