Torsion Box with MDF

[Rick Hickman]

Afternoon all,

Im thinking about building one of the fold down workbenches like April Wilkerson built a few years back. Have it fold down overtop of my lathe. Got me looking at torsion boxes (I know they're heavy, especially in MDF), and I was wondering why people DON'T build them this way;

Why not rip MDF strips and instead of standing them on edge, just lay strips on top of each other to make up the desired "height", face to face to build up the framework between the two skins of the bench? The faces are smooth, MDF is a consistent thickness, and if your width is off (as mine would likely be with my little jobsite saw) it won't affect the flatness of the bench.

In my mind you wouldn't even need to glue these lattice pieces together, just drill them for a dowel in two spots, and have the dowel go through the layers like guide pins. I cant see them shifting while you glue the MDF skins on. Alternatively, you could just countersink and screw the layers all together without much issue, just a whole heck of a load of screws. The countersink will stop the heads from throwing off the faces. Am I missing something or does this seem like a reasonable idea for someone without access to a nice cabinet saw/slider, or friend with something similar

Not trying to open an argument to be clear, just genuinely curious because edges of MDF suck, faces of MDF are better and it is such a consistent thickness that it seems reasonable this would work well.

Rick

[Jim Becker]

Laminating the partitions like you describe requires a "yuge" amount of effort compared to ripping and I don't think it's going to do anything more for the structure other than add weight and complexity to the build. What's more important is finding a way to build the torsion box on a flat surface so that it, in turn, ends up perfectly flat. The edges of MDF are perfectly fine for glue-up after cutting and smoothed off with a quick stroke or three with abrasive on a flat board.

[Cameron Wood]

If you took a stack of MDF rips and laid them across sawhorses, they would deflect under their own weight, let alone pushing down with a finger or two.

A rip of MDF on it's edge, not so much.

[Jamie Buxton]

No, torsion boxes can be built so that they are quite light for the strength they provide. I have a torsion box which is 72x42x4.5”. It weighs 35 pounds. If I support it at the ends, and sit in the middle, it deflects less than 1/32”.

The spacer grid is a 1/8” plywood on 6” centers. The skins are 1/4” plywood.

[Brian Runau]

Like others have said MDF strength is on it's side not face in this application and it would weight how many times more due to laminating to get the height. Brian

[Richard Coers]

Who wants to do a lot of rips of MDF? Do you want to test out your dust collection? Filling the box with wider rips of MDF eliminates one of the big benefits of a torsion box, light weight. 1/4"plywood internal ribs for the torsion box work great!

[Chris Parks]

The whole ethos of torsion boxes is light weight. I built a few of these in long lengths, the internal grid was 3mm MDF and they are very stiff.

[John TenEyck]

Gluing the faces onto a stack of unglued pieces provides no bending strength since it allows the unglued layers to slide over each other. The grid can be very thin and light, as said by many. Hollow core doors use corrugated paper cores. The key is the core must extend from one skin to the other, uninterrupted, in order for the glued up unit to be stiff.

As long as we're on the subject, note that the deflection of a beam, of which a torsion box is a type of, is:

Deflection = FL/4Ebh^3, where:


F = force
L = beam length
E = Young's Modulus
b = beam width
h = beam height (thickness)

So what's this tell you? It tells you that deflection is inversely proportional to thickness to the third power. A 2" thick torsion box will have 1/8 the deflection of one 1" thick. It also tells you that using thick skins offers no benefit with respect to deflection for 2 torsion boxes of equal overall thickness.

It also tells you that using materials with a higher Young's Modulus will result in lower deflection. Steel 2 x 4's for the internal grid, for example, would have less deflection than one made with wood 2 x 4's. The E of steel is 200 (GPa, but the units are irrelevant here), while Douglas Fir is 13, and MDF is 4. So a torsion box made with steel 2 x 4's will have 15X less deflection than one made with Doug Fir for the grid, for a given load placed on it, as long as the grid is well bonded to the skins, as always is the case with torsion boxes.

John

[Chris Parks]

So what's this tell you? It tells you that deflection is inversely proportional to thickness to the third power. A 2" thick torsion box will have 1/8 the deflection of one 1" thick. It also tells you that using thick skins offers no benefit with respect to deflection for 2 torsion boxes of equal overall thickness.

It also tells you that using materials with a higher Young's Modulus will result in lower deflection. Steel 2 x 4's for the internal grid, for example, would have less deflection than one made with wood 2 x 4's. The E of steel is 200 (GPa, but the units are irrelevant here), while Douglas Fir is 13, and MDF is 4. So a torsion box made with steel 2 x 4's will have 15X less deflection than one made with Doug Fir for the grid, for a given load placed on it, as long as the grid is well bonded to the skins, as always is the case with torsion boxes.

John

The TB's I built were about 3 metres long and the skins were 6mm MDF, there are photos somewhere on this site. It always astonishes me that people don't want to believe the rigidity of TB's using minimal thickness materials and then build using 16 or 18mm skins and similar internals. It is the depth of the internal frame that makes them rigid not the material thickness.

[roger wiegand]

We had an enormous conference room table (like 6X18 ft) that was made with veneered quarter inch MDF on a 2" thick matrix of honeycomb cardboard. It was incredibly stiff, deceptively light, and at least as judged by things rolling around on the surface, flat. I have to guess that it was vacuumed bagged on a very flat table. Building modifications done after installation made it necessary to cut it up to get it out when the company folded.