Torsion box

[sdakotadoug]

When I build, the floor will be 9X5 and I will be using a torsion box. Do I run the uninteruped wood (not the plywood) the long way and the interuped wood the short way or the other way around. I expect the floor to be 1/4 inch top and bottom and the stringers will be 3/4 material. I'm confused now are you. Anyway I hope you know what I mean, Thanks Doug

[Cliffmeister2000]

In cabinet and door making, the long pieces are the uninterrupted ones.

The long pieces are called stiles, and the shorter cross pieces are called rails. I don't know if this applies to torsion box construction, but it is universal in door and cabinet face construction.

[Rock]

Hard to tell in the pic but I made mine with solid members running crosswise:



Someone just asked this question (I think) so I looked up the reference - Andrew mentioned it in the awesome Ultralight trailer thread from which my trailer is made.

http://tnttt.com/viewto ... c&start=15

My floor is 1.5" framing lumber, 1/4" luan top and bottom, thickened epoxy as adhesive, staples, blueboard infill for insulation, covered top and bottom with fiberglass cloth and epoxy. My bench (made from hollow core doors and 3/4" MDF is dead flat so the floor is very flat as well (within 1/16".) As outlined in the Ultralight thread my trailer has no frame with the axle brackets and tongue simply bolted to this floor. Without mattress/electric this trailer weighed in at 400 lbs and is a (waterproof) rock. Hope this helps.

Eric

[Cliffmeister2000]

I think Andrew meant that many beams running the shorter span would be stronger than many beams running the longer span. I don't see where he addressed which should be uninterrupted.

[angib]

Do I run the uninteruped wood (not the plywood) the long way and the interuped wood the short way or the other way around

If you are building a sandwich panel (skin top and bottom) then I don't think it matters - the framing is there to space and connect to two skins to each other - their contribution to overall strength isn't that important.

If there has to be a 'right answer', I would run the shorter pieces continuously and interrupt the longer pieces.

I'm still trying to work out where this 'torsion box' name came from, as torsion has nothing to do with its use....

[Rock]

British cabinet maker Ian Kirby wrote an article over 15 years ago for a woodworking magazine where he built shelves exactly like my floor that he called torsion boxes. They had a nominal recess at the wall side that accepted a cleat screwed to the wall. There were no brackets or support other than the cleat. Not really much help but it's a reference going back at least that far. In this sense they are cantilevered and there is a moment (torque) around the attachment point (at the cleat.)

FWIW (not much) I also placed my continuous members crosswise because my tongue bolts to these and also the bulkhead divider between the cabin and storage area is attached to one (through the floor.) Kenny Ray's extensive stress analysis (which was WAY over my head) showed that this bulkhead sees a lot of stress. But in the end it likely doesn't mean a thing.

Eric

[sdakotadoug]

Thanks for all the replies. With the shortest ones being the uninteruped one I could probably use some 5 foot baltic birch plywood that I have lying around and save on the solid wood stuff. Doug

[DogStar]

A half lap joint on both directions would be the best. Don't ya think?

Eric

[kennyrayandersen]

Indeed you usually want the continuous beams going the short way. The bending stress and deflection are less; but, as was pointed out, in a bonded panel the skins react most of the load so it probably won’t make too much difference either way.

[Xanthoman]

I realize this is old but I’m case anyone happened on the question again...Torsion boxes are so named, I believe, because they resist torsion and remain planar and true without warping in any direction. As far as orientation of spliced pieces it would make most sense to put the continuous running length along the axis with the greatest bending moment (longest span typically). However, in a well made torsion box the joints are very tight and the skins hold the joints together and “see”, as mentioned previously by another, only a space separating the skins increasing the geometric inertia. The most important elements to remember on any composite structure such as torsion boxes or their single stringer counterpart, the I-beam, is that the web takes shear forces and the skins or flanges take the tension and compression forces. This only remains true as long as the interface between the flange and web is secured appropriately. So, make sure the skins are well adhered and it should technically be acceptable to have all sorts of gaps in the torsion spacer system. Admittedly, the unconnected points would be weak points where the skins could buckle if put under compression. So, in short, it shouldn’t matter your direction of continuity, but if there are gaps they should be on your tension side of the box as they have less effect on failure on that side of the system.

[Tom&Shelly]

I realize this is old but I’m case anyone happened on the question again...Torsion boxes are so named, I believe, because they resist torsion and remain planar and true without warping in any direction. As far as orientation of spliced pieces it would make most sense to put the continuous running length along the axis with the greatest bending moment (longest span typically). However, in a well made torsion box the joints are very tight and the skins hold the joints together and “see”, as mentioned previously by another, only a space separating the skins increasing the geometric inertia. The most important elements to remember on any composite structure such as torsion boxes or their single stringer counterpart, the I-beam, is that the web takes shear forces and the skins or flanges take the tension and compression forces. This only remains true as long as the interface between the flange and web is secured appropriately. So, make sure the skins are well adhered and it should technically be acceptable to have all sorts of gaps in the torsion spacer system. Admittedly, the unconnected points would be weak points where the skins could buckle if put under compression. So, in short, it shouldn’t matter your direction of continuity, but if there are gaps they should be on your tension side of the box as they have less effect on failure on that side of the system.

Very interesting! Thank you Xanthoman.

This confirms some of what I thought I knew, and adds to it. (An old electrical engineer with a semester each of "statics" and "dynamics" mechanical engineering survey courses from 34 years ago.)

They make optical benches using torsion box construction, and the ability to resist warping is undoubtedly an important reason why.

Tom

[Tomterrific]

As the glue joint between the long and cross members is small and therefore weak, cut a block of the same thickness and glue it into the corner attaching the two members with a larger glued area.

Tt

[QueticoBill]

I agree that the framing members and their splices are of little concern compared to the plywood skins. It's been quite a few years but I followed the American Plywood Association guide for stressed skin construction, now available on line I think. IIRC, the standard required overlapping splices of a foot - so a two foot piece of plywood of same thickness glued well under the skins.

[QueticoBill]

Its titled: Plywood Design Specification, Supplement #3

Described as: Design and fabrication of plywood stressed-skin panels. Third of five supplements to the Plywood Design Specification. Revised December 1, 2014.

Here: https://www.apawood.org/publication-sea ... spec&tid=1

You do have to register.

[rkanz]

My floor is a torsion box with 1/4” skins with 3/4” ply stringers. Made a very rigid floor.

The pink panther foam adds lots of additional strength.




Sent from my iPad using Tapatalk