Using a Clampin’ Guide and a Skilsaw with a planer blade in it, I squared the ends of the tops and scribed the corner radius—some kind of convenient can—and cut that with a new blade in the Bosch jigsaw. After ROSing the edges to 150 I rounded over top and bottom 5/16 radius. Often I’ll use a smaller radius on the bottom, but this thickness looks so strong that, if anything, I wanted to lighten it up a little with the radius. Then I used an OS to 220 on the top, edges and radius.
My finish of choice for the tops was waterborne poly. I selected Minwax’s Helmsman UV, mainly because I had the better part of a gallon in stock. Typically, in my understanding, UV products tend to be a little softer to accept the wood movement expected in outdoor applications. That makes this not the best choice for a desktop, but since it is really a computer surface (not a writing surface) I went with the MW product.
My application system was foam roller and foam brush, a two-fisted technique I learned from a boat builder who used it with epoxy coatings.
Note on foam roller covers: there are two kinds: those made with a longitudinal seam and those which are spiral-wound. I find the ones at the retail places tend to the the former, which leave a clear line in the finish when that seam hits your surface. The latter kind I find at my paint store. Slightly more cost, a few cents, and worth it.
Now to the otherwise obscure images above. These tops are big enough that I hadn’t room for them on sawhorses. I did have room for them on my 4×8 work table, but that left me leaning over to deal with part of the surface, and unable to see under the edges which were over the worktable.
The solution: A couple of lousy, er, lazy susan bearings, probably from captain chairs, given me by a fellow woodworker. I mounted them on the plywood extenders shown, then screwed them to the underside of the tops. Then I screwed the extenders down onto the work table. Bingo. I could roll material on the near half of the table, coat the edges and tip off the underside of drips, and rotate the table and repeat for the other half.
Final prep after 220x was to spritz the surfaces with water, let it dry, and then sand it back with 220 again. Smart, huh? Well, I rolled on the first coat of finish and the grain came up like spring weeds in an untended garden.
I sanded it back with 220 again on an OS. That seems really aggressive—I would not do it on lacquer—but it left it very smooth and took considerable material off judging from the white swarf.
The roller system worked well. There were never any obvious joints in the finish; I was able to keep a wet edge this way and control the drips on the underside. It was a nifty setup. The only downside was, there was nothing else that could go on in the shop. So I completed the finish over the weekend, coming and going in an epic of petroleum inefficiency which ultimately led to a completely acceptable result on the tops.
The last image shows the horizontal halogen light which really helped me take care of evenness issues as I was working.
6 coats altogether before I was satisfied. It is not as flat as a solvent-based finish would give you. The benefit was, no odor for my neighbors to endure (as in lacquer) and no long open time for dust to collect on the surface (as with solvent polyurethane). I am comfortable with the tradeoffs.
Next: final assembly of the components.
-- "...in his brain, which is as dry as the remainder biscuit after a voyage, he hath strange places cramm'd with observation, the which he vents in mangled forms." --Shakespeare, "As You Like It"