Home Articles Forums Gallery Marketplace Videos

WoodCentral Forums

Knowledge You Can Build On — Est. 1998

Identify and use hundreds of woods

On Torsion Boxes, Yet Again

Posts

On Torsion Boxes, Yet Again

#1

On Torsion Boxes, Yet Again

John McGaw

There have been numerous discussions about the benefits and possible pitfalls of torsion boxes over the years. A recent one touched upon using foam as a filler material rather than the traditional wood ribs and stretcher approach. I did an experiment with a foam-cored mantel 'slab' for my downstairs remodel. I needed a 90" X 9" shelf for the mantel over the brick fireplace and, there being no good structure to mount it, a lightweight fake would be best.

What I made used 2-1/2" X 3/4" poplar stock exactly as it came from the rack at the local borg, selecting the straightest (yeah, right) pieces I could find. These were cut to size, rabbeted 1/4" deep to suit cheap plywood and 3/8" wide on both sides. A 90X9 box was biscuited together and glued up with mitered corners. I then glued hardwood plywood which the borg sells for floor underlayment into one side's rabbets using regular yellow glue, added nearly every clamp I had and then dried it overnight.

I had some big slabs of semi-rigid packing foam which my kitchen sink was shipped in and, while not nearly so rigid or strong as normal insulation foam, I figured that it would do and it had the great benefit of being free. The foam was a bit thick to fit into the proposed space so I ran it through my thickness planer and then cut it to width and length to fit the recess in the box perfectly. The foam was glued into the box using West System epoxy, mostly to the plywood but as best I could to the box sides too. I smeared some more epoxy onto the top of the sandwich, added yellow glue to the rabbet and then added the other plywood skin and then clamped again overnight.

Some of the seams weren't quite perfect so I filled everything with auto body filler to make sure that I would have the look of a perfect slab. After a lot of sanding I did a minor roundover on all exposed corners and edges.

So, how strong is a foam torsion box? Supporting the box between sawhorses at the very ends I took the plunge and sat in the middle. Not only didn't I plunge, the deflection was nearly undetectable, certainly < 1/4" so I can say that a 90" X 9" X 2-1/2" torsion box with 1/4" plywood skins can easily support a 195-pound dead load and would certainly be up to much more. This isn't too shabby considering that the whole assembly weighs just over 13 pounds and was made with cheap or free materials.

I still need to design and make some corbels before I can install it but I'm certain that it will last longer than I will and will hold up whatever little things it might need to.

Re: On Torsion Boxes, Yet Again

#2

Re: On Torsion Boxes, Yet Again

Alan in Little Washington (NC)

Yup, that works well for a torsion box.

I was building a two place experimental class aircraft years ago (based on a Rutan design) that used wings made with hot wire cut foam cores, specially oriented, multi-layer, fiberglass skins, and a small fiberglass shear web just forward of the aileron slot. The wings were supposedly rated for 6 g's!!!!

Re: On Torsion Boxes, Yet Again

#3

Re: On Torsion Boxes, Yet Again

John McGaw

The only bummer for the project was that the only bulk epoxy I could lay hands on quickly was West's T-88 structural which is uber-expensive. There is probably something cheaper that would have served my purpose but I was in a hurry to use up my free junk materials.

Re: On Torsion Boxes, Yet Again

#4

Re: On Torsion Boxes, Yet Again

Barry Irby

I like it, Great save. Something I would do.

My question would be wanting to know how much the foam added to the project?

For a 9 by 90 by 2 3/4 Mantel or shelf I would have built it with a top and bottom and perimeter frame and a rib down the center the long way, dividing it into two 4 1/2 by 90's. I might have put in a few cross pieces. Yellow glue. OK, so your next project is to build one that way and sit on it and check the deflection. Let us know. ;)

I am turning into an old Fuddy Duddy and reluctant to try new fangled stuff. I wonder about the glue and the foam and the heat coming from the fireplace.

OTOH, if you could find a local supplier of SIP's (Structural Insulated Panels), you could have used a rip off one of those. They are basically foam between two sheets of plywood. Incredibly strong and they are building buildings out of them.

Re: On Torsion Boxes, Yet Again

#5

Re: On Torsion Boxes, Yet Again

John McGaw

This is something I've wanted to try for a long time and this proved to be the needed opportunity. The whole idea of 'something incompressible between tough skins' is well proven in many fields from hollow-core doors to racing yachts so I was certain that it would work, just not how well. I don't figure that heat will be a problem since the slab is quite a long way from the firebox opening and the the fireplace draws very well.

After I get done rebuilding a bathroom I want to add a very long single floating shelf to the gallery/library room. We're talking 16'X8"X1.5" so light construction is a must. Right now I'm considering a sandwich of 1" rigid foam with 1/8" hardwood plywood skins with the front and rear and end wood being added after the sandwich is made by cutting dadoes into it just a one would with solid wood stock. If I make it in two 8-foot sections with a very precise slip-together joint in the middle I figure I can get away with it. It should weigh next to nothing so supporting it should be pretty easy so long as I don't want to display my soapstone turnings and carvings.

Re: On Torsion Boxes, Yet Again

#6

Use the correct foam

Alan in Little Washington (NC)

Not all foams provide the necessary stiffness/resistance to compression. The best place to check for foam is Aircraft Spruce and Specialty Company. They have been selling foams and epoxy for experimental aircraft construction for over 30 years; since the early days of Rutan.

While different applications impose different criteria, the ideal foam is rigid with a tight, open cell structure, like used to build surfboards- NOT STYROFOAM. A hot wire saw can be used to cut many of the good foams easily and accurately into curvaliner shapes, like those needed in a wing airfoil.

Since too much epoxy can add unnecessary weight (especially important when building an aircraft) the common practice has been to fill the open cells at the surface of the foam with a slurry of epoxy and glass micro-ballons (mixed to a peanut butter consistency) before applying the glass fabric. Once the glass has been applied and fully wetted out with epoxy, it is also common practice to squeegie out air bubbles and any excess epoxy.

To do this often requires a slow-setting epoxy that doesn't generate a lot of heat when it cures. It also requires a fairly viscous epoxy. When doing large layups we would heat the epoxy which made it more viscous and easier to apply and wet out the glass fabric.

Polyester resin used in boat building should not be used. It does not bond nearly as well as epoxy, kicks off too quickly, generates too much heat (can actually melt the foam), and actually dissolves many types of foam!!!

Re: On Torsion Boxes, Yet Again

#7

Ellis Walentine

A word on cantilevered shelves

Ellis Walentine

Back when I was making a living as an architectural woodworker, we made lots of cantilevered torsion box shelves. I also worked with Ian Kirby on one of his well-known articles on torsion box construction, where the lead photo is of him sitting on one of his cantilevered boxes. Here are a couple thoughts I've distilled from those experiences...

First, the height (thickness) of the box is directly proportional to its resistance to deflection in cantilever. It is akin to the width of a rail in a table apron, where a taller shoulder provides more rack resistance.

Second, in another mortise-and-tenon analogy, the tighter the edge of the box is held to the wall, the stronger the resistance to deflection. (This is similar to a mortise and tenon joint, where the glue strength of the tenon to the mortise cheeks is what gives the shoulders their utility. If the glue is flexible, or the shoulders not tight to the stile, racking can more easily occur.)

Third, it is important to maintain a rock-solid interface between your wall cleat and the wall itself. Drywall, in particular, is not solid enough to bolt through. You need to have a virtually incompressible slip of wood or other solid material between your wall studs and your wall cleat. These spacer blocks should be the full height (thickness) of your torsion box.

Ellis

Re: On Torsion Boxes, Yet Again

#8

Re: On Torsion Boxes, Yet Again

William Duffield

A minor point, but it might help someone looking for the uber-expensive stuff: T-88 is a System Three product, not a West System product.

Re: On Torsion Boxes, Yet Again

#9

Re: Use the correct foam

Tom

Cool post. Just to add:

"Not all foams provide the necessary stiffness/resistance to compression."

Very important point. And these strengths need to be matched to the intended service and other player like the fiber and resin system.

" The best place to check for foam is Aircraft Spruce and Specialty Company. They have been selling foams and epoxy for experimental aircraft construction for over 30 years; since the early days of Rutan."

It is just one of many places, boat shops are often more locally available to pop centers. Insulation houses, and local building centers.

"While different applications impose different criteria, the ideal foam is rigid with a tight, open cell structure,"

The strongest and best foams are closed cell, though many open cell foams are good, like divinicell. One problem with open cell foams is the resin issues you raise below, and what happens to them if they get wet.

Kraft paper and balsa core are also good. Balsa is probably the strongest of all, but also heavy, though light compared to other woods. Often egg crating with webs works out lighter, and you can make it yourself. You could make your own "balsa" out of another light material, just remember to prepare it like a butcher block with the grain running between the outside panel, not end to end.

"like used to build surfboards- NOT STYROFOAM. A hot wire saw can be used to cut many of the good foams easily and accurately into curvaliner shapes, like those needed in a wing airfoil."

Actually those are made of styrofoam, and some people make decks of styrofoam, like the L7 trimaran, not my cup of tea though. I made my daggerboard out of triaxial over blue styrofoam precisely because I wanted to hot cut it. Most of the quality foams have very nasty things in them that you don't want to hot cut. And the manufacturers specify not to do so.

https://www.google.ca/search?q=styrofoam+surfboards&client=firefox-a&hs=wQW&rls=org.mozilla:en-US:official&tbm=isch&tbo=u&source=univ&sa=X&ei=duTLUoG0Noqy2QWTloHQBg&ved=0CFQQsAQ&biw=1024&bih=625

That said you could could make a much better board if you used something other than styro, and now that CNC shaping is in, they can cut the hard foams, and they play better with the tougher fibers like carbon.

"Since too much epoxy can add unnecessary weight (especially important when building an aircraft) the common practice has been to fill the open cells at the surface of the foam with a slurry of epoxy and glass micro-ballons (mixed to a peanut butter consistency) before applying the glass fabric. Once the glass has been applied and fully wetted out with epoxy, it is also common practice to squeegie out air bubbles and any excess epoxy."

Good info. The epoxy weight might be a the lesser problem, than the fact the epoxy can surprise you by either getting sucked right out of the joint, or conversely the air in the core expands into the skins.

"To do this often requires a slow-setting epoxy that doesn't generate a lot of heat when it cures. It also requires a fairly viscous epoxy. When doing large layups we would heat the epoxy which made it more viscous and easier to apply and wet out the glass fabric."

Few epoxies actually give off much heat when put in a thin coat. Your point is correct, but it won't discount any normal cure range epoxies from consideration , be they marked slow, medium or fast.

"Polyester resin used in boat building should not be used. It does not bond nearly as well as epoxy, kicks off too quickly, generates too much heat (can actually melt the foam), and actually dissolves many types of foam!!!

"Certainly wouldn't bond with it. I have been using epoxy pretty much exclusively since the 70s, Poly has a long history of use also with foam. Vinyl ester seems to be the favourite today if one isn't gluing to wood.

Re: On Torsion Boxes, Yet Again

#10

Re: Use the correct foam

Alan in Little Washington (NC)

"like used to build surfboards- NOT STYROFOAM. A hot wire saw can be used to cut many of the good foams easily and accurately into curvaliner shapes, like those needed in a wing airfoil."

Actually those are made of styrofoam, and some people make decks of styrofoam, like the L7 trimaran, not my cup of tea though. I made my daggerboard out of triaxial over blue styrofoam precisely because I wanted to hot cut it. Most of the quality foams have very nasty things in them that you don't want to hot cut. And the manufacturers specify not to do so.



Actually, "Styrofoam" is a brand name. I should have said styrene foam (or polystyrene) which was Styrofoam's original product. Styrene foam, is a relatively soft, closed cell foam that you can break apart into small foam beads, and is still use today for foam ice chests, appliance cabinet sound/heat insulation, boogie boards, etc. Real surboards and most quality fiberglass boat products did not and do not use styrene foam. They used/use the stiffer urethane (polyurethane or similar) foam. The polyester resin used in surfboard and boat construction uses a styrene monomer catalyst that melts styrene foam! My brother and I found that out in the early 60's when he ordered a used "foam" but unglassed surboard thinking it was a blank. It turns out it was cheap "finished/manufactured" surfboard with steel rod reinforcing and a relatively soft, unglassed styrene body. We did a small test and found polyester resin quickly melted the foam so we were forced to use the much more expensive epoxy to glass it !!! A few years earlier a young California surfboard shaper named Hobie Alter teamed up with a guy named Gordon "Grubby" Clark, who developed Clark foam, to produce the very first of what would become the revolutionary new wave of foam/fiberglass surfboards, then the famous Hobie Cat- like the one that sits in my garage- they were/are made with urethane not styrene foam.

One of the lightest and most rigid materials to use for the web of torsion boxes, especially thin ones, and which is widely used in aircraft panels and interiors is, a honey comb- both fibrous, non-flamable (like Nomex) or aluminum. Ian Kirby explored its use in his seminal torsion box article.

I'm sure this is more than anyone cared to know! :)

Re: On Torsion Boxes, Yet Again

#11

I remember that photo of Ian Kirby in that article

Howard in Toronto

That's useful information Ellis. Thanks.

I had square metal tubing arms welded to bar stock that I screwed to the wall using butterfly connectors and just slid the shelves over the arms.

16 years later, what seems like tons of books and no deflection.

Re: On Torsion Boxes, Yet Again

#12

Re: Use the correct foam

Tom

Yeah, they use that nomex honeycomb in a lot of aircraft parts. I was once in a crash where the guy who ended up piled two above me had his head driven through the nomex honeycomb type composite they use to make the overhead storage boxes. I never saw his head until several weeks later in the hospital. I would not like to have to push my head through that stuff. I was surprised he was alive.

Re: On Torsion Boxes, Yet Again

#13

Re: Use the correct foam

Tom

A pal of mine down the road just built a SUP using the actual white crumbly stuff. He used my hot saw to cut it. then he shaped it. There was poly snow in the gutter for a year! How weak the foam is is only relevant to the load. I used the Dow insulation stuff in my centerboard, and it has worked out. It is used to make various aircraft also. Some use it for the webs in wings, others make the whole aircraft out of "aircraft grade blue foam". One being the Skypup

http://www.youtube.com/watch?v=Wj-mkv5Fq-Y

We ain't talking Boeing here!

Re: On Torsion Boxes, Yet Again

#14

Ellis Walentine

Attaching your bar stock...

Ellis Walentine

I'm curious as to what the wall material was, how tall your typical torsion box shelf was, and how those butterfly connectors were used. I tend to use them for drywall or plaster, and I'm wondering how a lever arm like your shelf would hold up to heavy loads without deflecting or twisting the connectors right out through the plaster?

I used to do a similar thing with 1/2" steel rods drilled right into the studs. That worked fine too. I needed to use a shop made alignment jig and a long aircraft bit to get the hole perfectly perpendicular to the wall.

Ellis

Re: On Torsion Boxes, Yet Again

#15

Best description

Howard in Toronto

I drilled holes in the bar stock. They're not evenly spaced because I wanted to hit as many studs as possible. Wherever I couldn't, I just drilled a hole and pushed in a butterfly connector.

The shelves are -

- 1/4" MDF skins

- 1 1/2 inches tall (1/4" for the top skin + 1/4" for the bottom skin + 1 inch ribs )

- 66 inches long

- open at their backs so they just slide onto the arms

- have multiple closely-spaced internal ribs to keep their shape in the absence of a back

The walls are the early 1970's version of drywall.

Howard

Re: On Torsion Boxes, Yet Again

#16

Re: On Torsion Boxes, Yet Again

Tom

T-88 is Chemtech, which System 3 bought out. I thought it was made in Canada, but apparently Ohio? S3 has been buying up other people's stuff, and generally raising the price of their offerings since they bought out S3. I like WEST for most everything, but torsion boxes work real well with any glue. Epoxy has been around in boats long enough that the old guard like the Gougeons, and Kern Hendricks, are gone.

Re: On Torsion Boxes, Yet Again

#17

Re: On Torsion Boxes, Yet Again

William Duffield

Jan Gougeon crost the bar about 13 months ago. He was only 67.

Re: On Torsion Boxes, Yet Again

#18

Coming to this thread late *LINK*

Keith Newton

It is great to hear about some of the depth of experience coming from some of you here.

Like most of you, I suspect,,,, I started out with the WEST brand, but got tired of the pump hassle, then switched to System3. But have been using FGCI for the last 15 or so years. I find their prices to be a lot more reasonable than either of the above. They also provide a good variety of cores, fibers etc. as well.

I'm not doing much boat work or anything dealing with laying up glass, so the one I use mostly around the shop, is a bit thicker than those for wetting out glass or other fiber. It is a 1-1 mix ratio, but is very forgiving by as much as 2:1 or 1:2, and will still kick. I tested that a few months ago, not that I suggest being that careless, but gives a little piece of mind, when just poring X amount into two cups. If you ever were distracted by someone asking a question while you were in the process of pumping , say,,,, 14 strokes, or was that 16? and loosing count, not to mention when one takes in a bubble.


http://www.fgci.com

Re: On Torsion Boxes, Yet Again

#19

Re: Coming to this thread late

William Duffield

I agree, measuring West System with the pumps is courting disaster. It's also wasteful (which makes it even more expensive) for many small jobs, like attaching handles or filling knots or small furniture repairs. What I do most of the time is either use a miniature digital gram scale or these graduated measuring cups:

http://www.leevalley.com/en/wood/page.aspx?p=20009&cat=1,110,42967

and very careful manipulation of the pumps.

You can mix 105 resin with either 205 or 206 hardener in a 5:1 ratio by either volume or weight. I've found that you can mix pigments and fillers from one manufacturer with resin and hardener from another. I color the epoxy with dry earth pigments, vegetable black and even TransTint dyes. Dry tempera powders also work. I suspect TransFast dry dye powders will work as well.

One lesson I learned the hard way is that it is necessary to completely mix the resin and hardener before adding any filler or colorant. Otherwise, the epoxy just might not harden.

Thanks for the FGCI link.