In mid-2006, I saw a beautiful coffee table at Crate and Barrel - for about $700. I looked at it and thought to myself, "I can make that," and decided I would do just that.
Funny how life has a way making you put off your plans. I stayed busy for the next year and a half making Mojolevers, but at the end of the riding season sales slowed down a bit and I finally had time to spare. I roughed out my design for a strong-but-light table based on a hollow-core design: 1/4" thick oak-veneer plywood sandwiched on top and bottom of a 1"-thick lattice of poplar, with the upper sheet covered by a 1/8" layer of oak parquet strips.
I knew I needed a truly flat surface on which to build this table. At first, I thought the concrete floor of my basement would suffice, but alas, it was exceedingly unflat, with variations up to 1/2" over the required 5-foot span of the table. What to do???
As it turns out, I'm not the first person to encounter this issue. In fact, a common answer for avid woodworkers is to build a "torsion box" work table. A torsion box is an engineering structure in which two thin skins are attached to an internal framework, forming a lightweight but very rigid object. It's commonly used in architecture, aircraft - and woodworking. I didn't realize it at the time, but my coffee table design was a torsion box!
Based on the pictures, main dimensions and recommendations from the above link, I went shopping and - with the aid of a good friend who owns a pickup truck - came home with 2 sheets of 1/2" MDF and one sheet of 3/4" MDF. MDF is a great material for this because unlike real wood, it's isotropic, i.e. there is no preferred grain direction, and so it's got very little tendency to warp.
The first job was to rough-cut the necessary pieces from sheets of MDF: 1/2"-thick sheet for the top/bottom surfaces and interior lattice, and 3/4"-thick sheet for the exterior side pieces.
A six-foot spirit level worked great as a known straightedge for guiding the circular saw. The cuts ended up being pretty good, but I purposely cut each interior lattice piece slightly oversized so I could run all of them through my table saw with a rip fence to cut them to the exact same width, guaranteeing they all stood up to the same level when glued in place.
There was a lot of cutting, and as you can see the circular saw spit out a lot of dust. I made a mistake in not implementing some kind of dust control, and ended up with the stuff all over the basement, requiring a LOT of cleanup after the fact. Lesson learned: next time I will hang polyethylene sheeting from the ceiling to build a temporary enclosure in which to do the cutting!
The next step was to glue up the subassemblies for the internal lattice. Each 6-foot-long shear web got fitted up with its share of short crosspieces, until I had six of the seven long shear webs (five internal, plus one exterior face) fitted with cross pieces:
This made the final assembly easier, giving me just a few moderately-sized pieces to handle instead of a whole bunch of little pieces.
Of course if you're going to build a work table that's nice and flat, you need a flat construction surface to begin with. These guys had an answer, but it involved using a planer and jointer - two rather expensive power tools - to create flat timbers. Not wanting to burn $1000 on power tools I would only use once in a blue moon, I opted for a cheaper solution: $80 worth of 2" steel tubes, which happened to be very straight right from the vendor.
The first task was to cut one of them into two 3' lengths, lay both on the floor and shim them both level and coplanar:
(in the background you can see the shear web subassemblies of the table laid out for a test fit.)
With the six-foot spirit level, it was a simple matter to check the level between all four corners of the support base. Once that was done, the remaining four 6' tubes got laid on top, with the level being checked along many lines in the X- and Y- directions, as well as diagonally:
Most of the tubes were dead-straight, but a couple of them had a very slight bow, about 1/32" at the middle of the span. By spinning them a 1/4-turn, I found that these tubes each had at least one side that proved to be dead-straight (or at least I couldn't see any light between the tube and the level when they were laid length-wise together). Presto, a level support surface on which to construct my work table.
I laid the top and bottom MDF sheets together on the steel tubes. The plan was to glue the subassemblies to the "bottom" sheet (which is on top); the "top" sheet was stacked underneath to provide additional support for the "bottom" sheet during my work.
Gluing down the first shear web subassembly:
A combination of clamps and deadweights held the shear web and crosspieces flat against the big MDF sheet while the glue dried. Each piece of deadweight (heavy steel tubes or angle iron) bridges only two cross pieces, ensuring that each cross piece gets weighted down against the sheet.
Working my way across the table, I was able to glue down two shear webs per night, one right after coming home from work and another just before going to bed. In this shot, I've also added the exterior face on the short side of the table (at lower left of image):
Gluing on the final exterior face on the long side of the table, with all of my C-clamps, Kwik-Clamps, and pipe clamps to hold it in place:
Finally, with some shuffling, my wife and I extracted the "top" sheet from underneath this whole stack, and then we glued it in place on top of the completed lattice:
I used all my clamps, along with a lot of dead weight. Each small box you see is a package of lead wheel balancing weights, about 5.6 pounds. Gallons of paint, 12" speakers, whatever I had. It didn't require heavy weights, just a lot of weights to cover such a large area; for the most part, the top sheet's own weight and lack of warpage helped to hold it down.
The top and bottom sheets were of course oversized by about a 1/4" on all sides. After the glue dried, I ran a router around the edges to cut them flush. Having learned my lesson from the circular saw/sawdust fiasco earlier, before starting I hung polyethylene sheeting from the ceiling to form a temporary enclosure around my work area, and then stuck the shop vac hose inside it to ensure a slight influx of air through all of the gaps in the sheeting. It worked beautifully: zero sawdust-laden air escaped my work area, making cleanup a breeze!
While building the torsion box, I kept thinking about what sort of base to build for it. For a long time I had tunnel vision, thinking of wood as the only possible construction material. Finally something clicked and I realized I had another option: the steel tubes that I was using as a construction surface!
While waiting for the glued-up table to dry, I drew up a design that would work with the five tubes I had bought - including the one that had already been sawed into two 3-foot lengths. Once the table was done, I bought some more hardware (wheels, feet, nuts, bolts) for the support base and cut the tubes into the required lengths:
After that it was out to the garage for some welding. I had to make the table in two pieces so that I would be able to get it through doorways and into the basement. here are the two finished pieces, with hardware attached:
The support base, assembled and ready:
At the far end of the base are two non-castering wheels; at the near end are a pair of feet, along with the two chrome handles on the crossbar. This configuration allows for easy transport by lifting one end, and yet helps keep the whole thing stationary when you don't want it to move. The feet are adjustable; if the tabletop is rocking, the feet can be lowered or raised as necessary to ensure support at all four corners.
Before laying the torsion box on top of the support base, I added pieces of heavy-duty furniture felt:
Finally, a torsion box work table, ready for use:
The torsion box itself is unbelievably rigid. The top surface is flat to within about 0.020" (that was the largest feeler gauge I could slip under the spirit level at any point), and even when the torsion box is supported at two opposite corners by a badly-leveled support base, this does not change perceptibly.
Despite its rigidity and size, it's surprisingly light weight. The torsion box itself weighs in at about 120 pounds, and the steel support base is only about 50 pounds. Enough mass to provide good stability while working, but it only requires a lift of about 85 pounds when picking up the handles on one end to move it around.
I plan to go ahead with it like this for construction of the coffee table. At some point in the future I'd like to cover it with Formica or something to provide a tough, glue-resistant surface - but for the time being, I'm pretty happy with it.
Questions? Comments? Email me!
©2008, Mitchell P. Patrie