Stressed Skin Bookshelves

[John Baum]

I would like advice about the stressed skin approach to creating some additional bookshelves for a bookcase built from walnut stained birch face plywood. The shelves need to be 13" wide and three different lengths, 24", 30" and 48". I would like to approximate some existing 1" thick shelves made from 1" birch plywood. I have two different types of "1/4 inch" birch face panels. One is 0.190" thick and has birch on one face, lauan on the opposite face and a ply of lauan as the core. The second is 0.235" thick and is birch veneer on both faces of a mdf core. I have a fair stock of solid birch trim from which I can create the front faces of the shelves. I envision dadoing the internal edges of the front (and perhaps other faces) to accept the "1/4 inch" birch facing pieces. I have a fair stock of recyclable redwood, nominal 1 x 3 which I could rip to use for webbing.

I'd like opinions about whether or not the stressed skin approach is reasonable for bookshelves, and, if so, what kind of webbing distribution would be reasonable for the purpose.

Thanks,

John

[Jamie Buxton]

What you're talking about is often called a torsion box in woodworker lingo. Use SMC's Search function to find lots of threads about them.

Unless you really need the light weight that a torsion box provides, it might just be easier to make solid shelves. You're talking a shelf that is two quarter-inch layers separated by only a half inch. Get some cheap 1/2" core material -- say 1/2" OSB -- and glue it all together.

I'd be cautious about using MDF for the stressed skins in a torsion box. For a given load, MDF stretches more than lumber (along the grain), so a MDF-skinned torsion box is going to sag more than one made with plywood. I'd also suggest running your plywood so the grain direction of the core is the long direction of the shelf, again for stiffness. That may be contrary to your esthetic choice, but it would provide a stiffer shelf.

[John Baum]

Thank you, Jamie, for suggesting the use of osb and confirming an alternative I considered. I was put off when I assumed that I would need to sand it to get a good glueline. Am I wrong?

When I worked at MacMillan Bloedel Research in Vancouver (1967 to 1970) the team in Building Materials occasionally had some face sanded sheets of Aspenite around. They were pretty in their own way. The Marketing Department was adamant about not letting people see it. They didn't want the product to get a reputation as a niche market art material when they were working hard to get it accepted as construction underlayment. As I recall, its real forte was its dimensional stability. Even multiply plywood is not as anisotropic; there is always one more layer running running in the face direction than in the core cross-ply direction, so it moves more in one direction than the other. OSB is completely random, or, at least, Aspenite was in those days.

John

[Chris Fournier]

I don't believe that OSB is totally random as the acronym stands for Oriented Strand Board. When you look at it carefully you can see that the "chips" are indeed oriented along the 8' axis - at least on the face layers, I can't recall if the inner layers might not be "cross-banded".

[Mike Nolan]

You do not need a good glueline when you are gluing two surfaces together, the shear stress is very low.

[Chip Lindley]

You do not need a good glueline when you are gluing two surfaces together, the shear stress is very low.

I would say that two surfaces such as the bookcase shelf in question would be just a bit stronger when both layers are completely coated with (yellow) glue and clamped solidly together. The glue will lend some rigidity to the shelf when cured; even moreso if ply is glued to both top and bottom faces. If the shelf can be flexed to a slight convex arch (pre-stressed) while clamped up, it will equalize to flat with a load of books applied.

I have a full set of Encyclopedia Britannica on a 36" shelf, made thusly. It sags about 1/8" during the year. I flip the shelf during "spring cleaning" so it can begin to sag in the other direction for 12 months. The 1/8" sag is not discernable unless I point it out.

IMO, a 1" thick shelf is much too thin to benefit from "torsion box" technology, unless it will only hold knick knacks; not books. Somebody prove me wrong now! ;-)

[Aaron Hancock]

. Even multiply plywood is not as anisotropic; there is always one more layer running running in the face direction than in the core cross-ply direction, so it moves more in one direction than the other. OSB is completely random, or, at least, Aspenite was in those days.

John[/QUOTE]

I believe you mean plywood is not isotropic, not anisotropic.

[John Baum]

Quite so, Aaron.

The wife and I went into the garage this afternoon and found a couple of shelves a neighbor could not bring himself to send to the dump. They are 0.70" 9-ply birch with a paper-thin finished maple skin. She undertook to sand the finish from the side that suffered nail pullout damage when the shelves were removed. We will glue some of our birch face lauan 3-ply to that face to give us a 0.89" thick shelf that should be plenty stiff enough for a ~30" span. Since they are 1.25" too narrow, I will glue and dowel a 3/4" birch trim strip to the exposed front of the shelf and glue and finish nail a 1/2" birch trim strip on the back to get to a full 13" wide shelf. The facings should improve the stiffness a bit as well.

Thanks to everyone for their contributions.

John

[Alan Schaffter]

IMO, a 1" thick shelf is much too thin to benefit from "torsion box" technology, unless it will only hold knick knacks; not books. Somebody prove me wrong now! ;-)

Depends on the material.

Years ago I was building a composite (Rutan) airplane from foam and fiberglass- wings, fuselage, bulkheads, etc., essentially every part was a torsion box. The flat bulkheads were made from dense, but lightweight, 3/8" thick white foam laminated on each face with fiberglass and epoxy- again a torsion box. Yes, the bulkheads were very, very light . . . but very, very strong!! I kept a 3' long by 12" wide piece in my shop as a demonstrator. When visitors would question my sanity after I told them I planned to fly in this foam and fiberglass airplane I built myself, I would take out my bulkhead demonstrator, support it between two cinder blocks, stand in the middle of it, and bounce!!!

So it is very possible to make a thin, lightweight, but strong torsion box!

A 1" torsion box with wood veneer over fiberglass, Kevlar, or other skin with a foam or honeycomb core would work great- by the way, that is how they make the fancy wood panels in the Gulfstream V and other executive jets. Here is more info. (http://www.bellcomb.com/materials.aspx)

http://www.bellcomb.com/UserFiles/Image/Exploded-Honeycomb.JPG

[Chris Friesen]

For shelves that will carry weight you actually want as much grain as possible running from side to side. Thus, solid wood is the best bet, followed by plywood/osb, then mdf and particleboard.

Punching some numbers into the sagulator (http://www.woodbin.com/calcs/sagulator.htm) it looks like plywood/osb might possibly be able to take the weight, but you'd be substantially better off with solid wood. Even poplar would be better.
(http://www.woodbin.com/calcs/sagulator.htm)