I have a project that incorporates long shelves that are supported only at the ends. I am thinking about using a veneered torsion box design to prevent sagging. Anyone have any experience using honeycomb cores for torsion boxes. I want to keep the shelf thickness below 1.25 in so I was thinking about 1/4 or 3/8 skins with a honeycomb core, solid edging and venner on the top and bottom. Any ideas or advise would be greatly appreciated. I have vacuum capabilities for clamping the assebly together.
Tom
Douglasville, GA
Replies
What resource are you trying to conserve by using torsion box construction? For any given thickness, solid wood will be stronger and stiffer than the torsion box.* At 1-1/4" thick it's hard to imagine that the saving in materials, compared to solid wood, will repay the extra fabrication time.
* Edited to add: Assuming your torsion box skins have about the same physical properties as the solid wood shelf you're comparing to.
Edited 1/20/2004 10:19:44 PM ET by Uncle Dunc
Unc -
I would think weight would be the primary consideration given the thickness of the shelves and, considering the comment "long", the overall span. The dead load of a solid wood shelf itself would be enough to produce a noticeable sag in the shelf. I once used 1 1/8" T&G plywood subfloor material to make some long span shelves, with plastic laminate and hardwood edge banding. These were bookshelves for my office. Plenty of heavy weight reference type books being supported. I had to regularly remove all the books (about once a year) and turn them over to reverse the bow they would develop.
I'm not sure if a veneer over a honeycomb core would be stronger than solid wood - just that for any given thickness, there's a limit to the span.
...........
From Beautiful Skagit Co. Wa.
Dennis
Tom -
We could offer more constructive opinions if you could provide more information -
What is the total span? What are the shelves supporting? How deep are the shelves and are they supported along the back?
From Beautiful Skagit Co. Wa.
Dennis
I've built torsion boxes using cardboard honeycomb core. I've built lots more with wood-grid core. In my opinion, the plywood grid is better. You can make wood grid quickly in your own shop, without special-ordering stuff from the other side of the country. You can make plywood grid to be any dimensions and thickness; cardboard cores offer limited choices.
Tom, you didn't mention the length, also the load imposed on the flat shelves. Try to 'Arch'up the shelves and only store objects with concave bottoms. Stein.
The longest clear span for the shelves will be 5-6 ft. I don't know if they will be attached to the wall for the whole length or not. I would prefer that they are but need to look at the wall construction first. As far as using solid wood, I am concerned about cupping and bowing after they are installed. The shelves are to be seperated by aluminum tube at only two or four points per shelf depending on whether the back is attached to the wall.Tom
Douglasville, GA
Hide 1/4" thick steel plate inside the front and back edgbanding. Groove the back of the solid wood banding and epoxy in the steel. I've done this on torsion box construction.
Tom:
Your requirements sound like an ideal application for a torsion box. I am not sure I agree with UncleDunc's comment about solid wood being "stronger and stiffer" than an torsion box. The physics of a torsion box ensures superior stiffness and rigidity for any given thickness.
I would go with the 1/4" skins. And I found http://www.mcmaster.com a good source for various thicknesses of core material.
I've built big 9' long countertops for a reception area this way (2" thick honeycomb core, 1/2" skins) and supported it only at the ends...it does not sag no matter how hard I try. The vacuum press will be indispensable for this.
>> I am not sure I agree with UncleDunc's comment about solid wood being "stronger and stiffer" ...
OK, I'm willing to learn. Tom proposed a 1-1/4" shelf made with 3/8" skins and a 1/2" hex core. If I start with a piece of 1-1/4" solid wood, slice 1/2" out of the middle of it, and replace it with hex core, how does that make the shelf stronger or stiffer?
The core in a torsion box has two functions, keeping the skins the right distance apart, and keeping them from sliding past each other. I contend that solid wood does both of those jobs just as well, probably better, and with no glue joint to potentially fail.
My understanding is that the virtue of a torsion box is that it gives you more strength and stiffness than a solid wood shelf with the same total thickness of wood, not the same overall thickness. Tom's proposal with 2 skins @ 3/8" will certainly be stronger and stiffer than a 3/4" solid wood shelf, but I don't see how it can be stronger and stiffer than an 1-1/4" solid wood shelf, assuming similar materials.
Same with your contertop example. I'm sure your 3" assemblies are wonderfully stout, and much cheaper than solid wood, but it's not obvious to me that they're stronger and stiffer than solid wood. Not that I expect you to buy a 3" chunk of solid wood for comparison testing. :o)
The thickness of torsion boxes is what makes them work. The farther you can move the skins away from the neutral axis, the more work they can do. The overall thickness limit is what made me question the fitness of a torsion box in this application. If Tom could live with 2" shelves, I'd say go for it, but at 1-1/4" I think a torsion box will have no mechanical advantage over solid wood, and very possibly no cost advantage, either, if you count the time that goes into putting them together.
Edited 1/21/2004 5:04:56 PM ET by Uncle Dunc
Unc. -
My reply was not mindful of the 'skin' thickness noted in the original design - 3/8". I was thinking more along the lines of a hollow core residential door with skins that aren't much more than 1/8" thick. And for that matter, they have a core made of, in most cases, cardboard on edge - not a plywood grid. And they're remarkably light and strong for what they're made from.
Thus my approach to the original design problem would be to maximize the depth of the interior grid filler, perhaps even going as light as 1/4" plywood, with 3/16" thick skins and a 3/4" x shelf thickness edge banding. I still haven't seen what sort of loading these shelves will have to support so that part of the equation will determine not just whether the shelves are solid wood but how thick they have to be. Torsion box or solid wood, there's a limit to the span, regardless. The torsion box elimnates a lot of dead load weight the shelf has to carry is one of the main advanatages I can see right off the top.
...........
From Beautiful Skagit Co. Wa.
Dennis
Your understanding of the torsion box is right on. I am sure you realize that in order to flex, shear force must overcome the bond of the glue area. It is the resistance to this shear force that makes the box work. With a rigid glue, this is not going to happen before catastrophic failure occurs.
There is also a much higher strength to weight ratio. I think this is the real advantage as the span increases. Given the same overall thickness, span distance, and load, the torsion box will flex less than a solid board. And remains so for a longer period of time. The bottom line is that a solid board is elastic...the fibers are elastic, no matter how thick the board. It must also support its own mass.
Granted not all torsion boxes are equal. I agree that there is questionable advantage to constructing a box as you've described, and that a torsion box's rigidity greatly increases with thickness. The core thickness must be maximized: Going to 1/4" skins will give a 3/4" core. Not too shabby...Better yet, use door skins...
But in the long run, I believe the advantage goes to the box.
I didn't specify 3/8 skins in my original post. I mentioned using 1/4 or 3/8 skins. I haven't deigned many torsion boxes so I am pretty flexible on the construction. I don't know what the loading will be. I have been told that it won't be much but we all know that intended usage and actual usage aren't always the same. For that reason I am interested in making the shelves as stiff/strong as reasonably possible.Tom
Douglasville, GA
Attach a cleat to the back wall and recess it in the shelf. Sorry I can't draw or post a pic. You could use foam panel instead of honeycomb. I've built some supported only at back and one side.
What type of foam panel do you use?Tom
Douglasville, GA
I probably should have qualified my statement to make it clear that the 3/8" skins was only one of two variants you had in mind, but I already have a problem with my posts getting too long. I don't think either variant makes optimum use of a torsion box's strengths, but it's true that I chose to discuss the one I thought was least optimum.
There is no doubt that a 1-1/4" torsion box made with 1/4" skins will perform dramatically better than a 1/2" think board the same length. And a 1-1/4" torsion box made with 3/8" skins will perform better than a 3/4" board. I don't think either torsion box will perform better than a 1-1/4" solid wood shelf, assuming the solid wood shelf and the torsion box skins are substantially similar in strength. But I'm not as sure of that now as I was before reading some of the other posts in this thread.
If the performance of the torsion boxes is no better than the solid wood, and if your time is worth anything, money wise, it might be cheaper to go with 1-1/4" solid wood, because cutting and finishing the solid wood will take a lot less time to than cutting, fabricating, and finishing the torsion boxes. And that time difference might favor the solid wood even if costs more than the materials for the torsion boxes.
I may be the only person who builds torsion box work benches with 3/8" faces. But I agree that solid wood shelves would be stronger than a torsion box of the same total thickness. And I agree that solid wood is the best solution for this project.
For 6' long shelves without back support 1-1/8" ply is too little. According to the APA Plywood is less stiff along its length than the softwoods it is made from. And those softwoods are less stiff than the hardwoods I use.
I would find some nice 6/4 hardwood and use that for shelves. Put a nice edge on the front.
>> Plywood is less stiff along its length than the softwoods it is made from.
That's what I suspected, since only half the layers run the long way. Microlam is another possibility. All the layers run the same direction. I don't know how it compares to hardwood in price or stiffness.
with a firm grip on my resolve and my courage screwed to the sticking place, I'm going to be heretical.
At 6 ft span and only 1.25 thick I'm thinking 1/4 in ply epoxied to a frame made of square aluminium extrusion. Front and back edges of the shelf finished in solid timber.
Ian
Ian,
I like your "out of the box" thinking! Sure makes sense to me.
Wayne
How about veneer on a hollow core door?..buy the door rip it, refill the cut edge, add veneer of choice..get the inch and an eighth closet door with no holes in it. they come as narrow as 12 inches.Go Stab yourself Ya Putz! Ya think I Parked here?
What about using a hollow box with a thin face and filled with a stiff but light polyurethane foam? This is a similar technique to that used by some manufacturers of extruded aluminium machine frame sections as a way of increasing both the rigidity and strength of long beams and works well.
Just my 2 cents worth
Scrit
Scrit
This is sort of what I'm think of. This is the construction of refrigerator panels. Not sure how well it will perform over time as it tends to delaminate when laoded. I was thinking along the lines of an aluminium extrusion (a scafold plank comes to mind) sheathed in wood.
The original poster was looking for something thinner than 1.25 in.Checking Sagulator, a solid wood 6ft book long shelf, 12 deep and 1 1/8 thick will initially sag MDF 1.46 inply 0.43 in Black maple 0.32 in White Oak 0.29 inTeak 0.33 inJarrah 0.27 inIpe 0.16 in
Sagulator calculates initial sag and suggests that the long term sag is 1.5 times as much. Sagulator also suggests that the visual limit is 1/32 in per running foot, so 6 ft equates to 3/16 in long term, or 1/8 in (0.125 in) initially.
None of the above falls within these limits. So I guess it's a torsion box or a timber sheathed Al frame
Ian
Perhaps you could compute the deflection for a torion box for the same loading.
I think you will find a wood core/ plywood faced torsion box to have more deflection than solid wood.
I believe a reasonable torsion box would require 2" of depth.
For those who wonder about these issues a 1-1/4"x12"x6' weighs about 30#. A torsion box might weigh about 10#.
A properly constructed torsion box will be dead flat and subject to less movement. A 5/4 x 12 would most likely cup and twist unless we're talking Q/sawn. Hardwood plywood usually comes already sanded to 180g. Torsion box construction would more readily lend itself to the use of hidden cleats. I could build the whole thing on site with portable power tools. Which I do,sometimes.
Thanks for all the input so far guys. Does anyone have a method of determining the size and strength of a torsion box?Tom
Douglasville, GA
Tom ---
Perhaps Scrit or Mike Rooney could do that for you. They both seem to make claims that I am unwilling to make.
On the other hand the APA web site has sufficient information if you read their documents carefuly.
take a 6' 8'' HC luaun door (cheap..) and support both ends on a coupla chairs..start loadin cinderblocks on like books..you'll be amazed..12'' door will hold blocks from end to end solid.
now recover that door with your veneer of choice..there is a book shelf.
Go Stab yourself Ya Putz! Ya think I Parked here?
I am leaning towards that very method. Would be easy to conceal a wall cleat that way as well.Tom
Douglasville, GA
Tom,
I'm a little late getiing into this, but since some have started to think outside the box, how about using some boatbuilding techniques? I worked for a few years building high tech boats and I think you could use a core of foam or honeycomb, skinned with fiberglass (like a surfboard), topped off with the veneer. It could be laid up all at once using vacuum bagging.
Just my 2 cents.
Richard
Edited 1/26/2004 6:39:46 PM ET by Richard
This forum post is now archived. Commenting has been disabled