Come on, do copper. The steel-core copper-jacket idea is great. Forget about twist/domino, here’s the solution: Mount the tubes to the floor and ceiling (and back as well if you want). More opportunity for using cool metal flanges and things!
What I don’t quite get about your design is how it will be stable horizontally. Take a pencil and stab it through a piece of paper. OK, it has tight tolerences as the hole exactly matches the diameter of the pencil, but if you pull the paper in any direction, it will still move the pencil off center and angle enough that over time, if applied to heavy books, would cause the whole thing to lean and then crash. You need to SECURE the dowels to each and every board. You could use strong wood glue, or maybe some type of wedge. Or maybe on the back side drill from the edge of the board into the dowels with long screws (which would be 90degrees from the direction of the cross-pegs underneath the boards, ideally).
But I’d still use steel rods with copper sleeves.
-Tcat
Like I said before: if you drill a hole in a solid structural part that exceeds one-third of the diameter/width/thickness of that structural part, you have compromised its strength. It will be much much more likely to fail in bending, compression, and tension. Almost all of the formulae for beam strength incorporate “cross sectional area” somewhere in them, and if you drill out half of the diameter of the wooden dowel, the cross-sectional area at the drilled point is greatly reduced.
I recommend steel pins to hold up the shelves, maybe 1/8" diameter. 1/8" steel pins will survive much longer under constant load than 1/2" wood, and their failure mode isn’t as catastrophic.
Eh. Inch-diameter copper pipe ought to work fine, but I’d still be concerned about the shear on the shelf-support pins because of the relatively thin wall. Stikes me that a better shelf-support design will be preferable if you use pipe. But if you’re going with dowels, it should be OK. (You could also probably use copper pipe flanges, which would be pretty strong, but that also strikes me as making this an extremely labor-intensive product) (Finally, I was looking through the design engineer’s bible - the McMaster-Carr catalog - and I saw they sell brass structural tubing, with fittings, (search for “brass structural”) which is made for exactly what you’re doing.)
First of all: you can see the support pins? Aren’t they underneath the shelf, out of sight? Second of all: a half-inch hole in an inch diameter dowel doesn’t leave a lot of wood around it for structure. It’s probably OK, but you’re significantly cutting into your factor of safety. Third of all, if you really want to get a copper look, do like some other people have suggested and clad the dowels with copper. You can buy thin copper sheets and hand roll them around a thin dowel to make a cylinder.
Some sort of bracing is absolutely positively needed. The wood is too soft to adequately withstand a sideways torque. It might feel tight after you build it, but I wouldn’t bet on it staying that way after years of loading. Your bracing doesn’t have to be cross-braces, though. Just something to prevent the upper shelf from swaying back-and forth. If your braces into the wall studs will prevent left/right motion as well as forward/back motion, then you should be all right.
One-inch dowels seem completely adequate to me. A half-inch hole seems large, though.
1/8" strikes me as small, considering that the weight transfer will be into end grain on the dowel. How about 1/4"?
First, let me reiterate how much we appreciate all your help and questions!
Horizontal stability
Here’s an experiment I just did: I took some corrugated cardboard (about 1/8[sup]th[/sup] inch thick) and cut it into rectangles, about 3.5" X 5.5". I punched four holes in it, and then slid in four pencils (The pencils are about 1/4" X 7.5" (The pencils were unsharpened, and are #2, like any good test taker). The spacing of the pencils is about 1" in from the width edge, about 3/4" from the length edge, 2" from each other along the width, and 3" apart along the length.
(fun with ASCII)
____________5.5"___wide______
___1" (from edge to first pencil)
_____________________________
| |
| o o |
| |
| o o |
|____________________________|
(I hope these are enough measurements/drawings to extrapolate position.
My ASCII skills take too long to try and put more in...)
I spaced the cardboard evenly, (about an inch between layers) and have a miniature, almost mock-up of our shelves. Ours will have six vertical supports, and of course the proportions are not the same (else I would have been able to easily fit six pencils), but I’m hoping to look at the general principle.
This differs a bit from the pencil-through-paper thought experiment above, as the cardboard has some depth to it. It’s now sitting on my desk, and as it just sits there it seems relatively stable. Of course, it’s just sitting there…
I put a bit of horizontal pressure on the tops of the pencils. The bottoms, the erasers, have enough friction to keep it from sliding/being pushed. Lo—it starts to fold sideways. But then it stops. It leans about ten degrees or so, then it comes to a complete stop. The only way I could push it further would be to rip through the cardboard. Of course, a few hundred pounds of books may do just that to our shelves.
But I noticed something. In order for it to move as it did, the cardboard shelf had to slide along the pencil. The side opposite the lean moved—albeit a very small amount—downward, while the side closer to the lean moved upwards.
Question: How applicable is this to the real world? That is, since each shelf will be prevented from downward movement, and the tolerances should be fairly high (again, we’ll be using a drill press and measuring the heck out of things) will the combination of 1" (the nominal thickness) boards, six vertical columns, and inability for downward movement result in a horizontally stable structure?
Beyond that:
For front-to-back movement, we will be able to place a brace just about anywhere on the shelves, so anchoring the shelves to a stud with a small bracket should not be a problem. However, there is no guarantee that the placement of the vertical supports will line up with wall studs. If the above isn’t safe (and pencil-and-cardboard experiments can only go so far), we’ll be attaching some sort of brace to the three close-to-the-wall vertical supports, probably a pipe clamp or the equivalent (since the top will have the most leverage, and be the least visible (due to both height and books), aesthetic concerns are slightly lower. If we put them through the sheetrock with a molly or other stronger anchor (i.e. not a simple plastic sleeve), will the combined strength of six attachment points (three supports, two screws each) suffice? Or should we find a way to use studs to support them horizontally?
Again—thanks!
Rhythm
I think you have definite issues with the copper idea…
You also have very definite issues with 1" pine shelving, intended to hold books, spanning 48" between supports. Check out this site: The Sagulator – WoodBin
The shear strength of pine if the pipes are allowed to wobble, as you described be an issue too. If you were using hard maple, it could be a different story.
One suggestion I have is to mount 1"x1" angle iron to the wall like this –> L and sit the shelf on the bracket. Screwing the pine through the bottom of the bracket. This will do a number of things; prevent movement, secure the shelves from falling forward if kids climb on them, help prevent sag in the shelf (don’t know if it will prevent it totally, .11" of sag over 48 inches is a lot) and it would take the load off of the back pins.
I would also add one more set of supports, breaking the shelf into three parts.
There, you’re essentially relying on the shear strength of whatever you’re using to attach the shelves to your dowels. Also, relying on a snug fit between the two means you have to deal with excessive amounts of torque if your shelf isn’t stiff enough-- tens of pounds (or more) of books with a couple feet of shelf acting as a lever on your inch-long joint. Over time, this joint will be racked back and forth and become weaker, not fitting nearly as snugly as before. I think this has the potential to fail rather catastrophically. If you anchor the back of the shelf to the wall, you’re essentially accomplishing the same thing as using diagonal reinforcing members across the back (the wall itself becomes the webbing), which solves this problem. Still, unless you’re also going to anchor each shelf to the wall, I’d personally be more comfortable with reinforcement on each shelf like the rabbeted strips that Rocketeer mentioned.
OTOH, I have a small bookshelf right here that has some sort of 3/4" thick hardwood shelves that are unsupported over a three foot span. The back is reinforced by a sheet of plywood, but it only reinforces the sides and top shelf. The shelf is certainly sturdy enough to handle a moderate amount of books.
What he said.