Diary of a Novice: Coffee Table
What follows is a chronicle of my table-building odyssey, including frustrations, mistakes, and finally, success.
First I had to figure out what kind of table I wanted. I knew we needed a coffee table at home, so that part was decided right away. Then I went around my house and looked at tables. With the exception of one round pedestal table, all of our tables were similar in that they included a top, an apron (which doubled as the support structure) and legs. A couple had drawers, but others didn’t. So now I had a plan. Table-top, apron or supports, and legs. I decided to forego the drawers because I had never done them before, and I didn’t need any more complications. I sat down in front of several tables to get an idea the height, length and width I wanted it to be. For style, I wanted something simple, casual, maybe a little rustic, but definitely unique to me. I figured I’d worry about the supporting structure later and I concentrated on the top.
Working directly on her workbench, White laid out the dressed boards in a way that was visually appealing to her. Once she was happy with the results she marked the boards.
A Quick Trim
A track saw, or in this case, a long straight-edge and circular saw, makes quick work of truing the jagged edge left after glue-up is complete.
Each leg was rabbeted on three faces, to accept the mating rails.
Once the inner and outer long rails were joined to the legs, White bored holes and added wood pegs, which were trimmed and flushed to the rails’ surface.
Inner Short Rail
With tape to protect against squeeze-out in place, the inner short rails brought the two sub-assemblies together for good. Screws and glue created a solid joint.
The outer short rail was mitred on both ends then glued and clamped in place, creating a clean look.
The top view of a leg joint, showing how the short and long rails mate with the leg.
First, the top
The table-tops we had at home were all boards of the same wood that had been laminated together to look like one solid piece. I ruled that out right away. I didn’t know if I had the gluing and grain-matching skills to pull that off. I decided I’d make something to accentuate the differences in the boards. Fairly quickly, I zeroed in on the cutting board idea, in which several different woods are glued together in different patterns and then sanded smooth. I decided I’d run the boards at a 45° angle to the side. I picked 45° mainly because it was marked on most of my tools and saws, and I wouldn’t have to do a lot of angle calculations. I also liked the angle idea because I’d never seen a table-top like that. Then I decided I would cut the strips in random widths without any kind of regular repeating pattern because it would be much easier to hide any cutting mistakes and would also be in keeping with the table’s somewhat oddball appearance.
All my ideas were roughly on paper or in my head. I hadn’t actually cut anything yet, so I hadn’t made any mistakes. Yet, that is.
Now I had to choose woods. For cost reasons, I decided I’d use as much as I could from my stash, but within those limitations, I wanted as much variety in colour and texture as possible. I had received my beautiful turned legs (more on them shortly) and they were walnut, so I definitely wanted a little walnut in the top. At home I had one board of cherry, one of walnut, one of poplar, lots of hickory and ash, and several little pieces of exotics. Then I spied some bits of spalted maple I had been saving that had come from my backyard. This spalted maple instantly became a must-have for this table. That was my first mistake, as you’ll see later.
My 4′ x 8′ workshop table has a sacrificial sheet of plywood on top, so I drew the outline of my table-top on the plywood, including 45° lines, to get an idea what it would look like. After drawing the table-top, it became obvious that at two opposing corners, the length of the 45° boards shortened considerably, so I knew that’s where I would put my exotics and maple.
It was clear that I was going to have to plane boards to the same thickness. This is when I first suspected the maple was going to be a problem. The maple was thin, about 5/8″. It had been cut and planed quickly by a carpenter buddy, and it was really too thin for this application. I had thicker pieces of maple, but they were bizarre shapes that were mostly warped, so I didn’t think I could improve on these nice pieces that were thin. The maple would be the limiting factor, and all the boards would have to be planed to match these thinnest pieces. If I had not been so determined to include the maple – which in the finished top did not even add up to 1 sq. ft. – my table-top could have been 7/8″ thick. Instead the finished top was less than 5/8″, which I would not recommend to others. I lay out the boards and got a rough idea of how many boards I would need and started dressing them to final thickness.
Then I made two mistakes in quick succession. Although I cut my boards longer than I needed them, they were often not quite long enough. When you lay out boards at a 45° angle, even if it’s only 25.5″ across my finished table, boards in the center are over 3’ long when laid at an angle. So when I arranged and rearranged the boards for glue up, more than once a board I wanted to use in a specific spot wasn’t long enough. So then I had to take out the planer again. And again. And again. You get the idea.
The next mistake was cutting the ends of the boards at 45° before planing and glue-up. I thought I was saving myself a step, and this would make my glue-up easier. Take my advice and leave the boards rectangular until the table-top is in one piece. Having already cut the angles, the boards were harder to manoeuvre through the planer, harder to run through the table-saw, and much harder to clamp during glue up.
One tricky glue-up
Glue-up started off straight-forward and then became more and more difficult because of clamping. I started with gluing two boards together, then four boards together, etc. Clamping was fine in the beginning, but the more boards I glued together the harder the clamping became because the boards were at an angle, and they wanted to slide back and forth. It is important to clamp across the seam between two boards. The more boards you are clamping at an angle, the harder it is to do it across the seam without the boards sliding. If you don’t precut the edges at the 45° angle, then you have “wings” on each side to help with clamping. I made a little jig for holding the square edge and clamping across. And remember that, when clamping anything long, you need long clamps. I had four long pipe clamps and wished I’d had more.
I wasn’t careful about cleaning up glue squeeze-out on the table-top and boy was I sorry. I had a hard time getting that glue off once it was hard and in one spot it actually stained my cherry. Scrape or wipe off as much squeeze-out as you can, while you can. Once it was all assembled I used a belt sander and an old 1/3 sheet sander to sand the table-top.
The belt sander takes stock off very quickly, but you have to have a good technique when using it. Also, sanding different woods of different hardness that are laminated together requires real concentration because the softer woods sand quicker than the harder ones. You have to hold a belt sander very flat and move it around a lot or it’s easy to sand waves and dips into your table-top. Don’t ask me how I know that. I sanded the table-top until I was satisfied with it, and finally used an edge-rounding plane to quickly remove all the sharp edges. Then I moved on to the table-base.
The table base
I wanted turned legs for the base, but I don’t have a lathe, or the skills to produce them. No problem. The walnut turned legs were made by Osborne Wood Products to the exact dimensions and style I was after. This allowed me to create a table that would have otherwise been impossible. My main preoccupation at this point was to not mess up the legs because they were beautiful. I spent a lot of sleepless nights worrying about those legs. I hated to cut into them but I knew I’d have to sooner or later. Since I’m a novice, I decided use very simple joinery. I used eight rails to tie the four legs together: four long rails and four short rails. The rails sit in rabbets cut into the tops of the legs where I have the rails glued and pinned (for the long rails) and glued and screwed (for the short rails). I used wood pins where they would be visible and screws were they wouldn’t be seen.
Since my rails were all 2-1/2″ wide and 3/4″ thick, I made three rabbets on each square leg top that were 2-1/2″ wide and 3/4″ thick. My exterior long and short rails have a mitred end so that they meet neatly and the structure under the table looks finished. I cut those small mitres on the skinny rails with a mitre box and was pleasantly surprised at its accuracy.
Rabbeting the leg tops was nerve wracking, to put it mildly, especially after I messed up the first rabbet on the first leg while trying to cut the rabbet with a dado stack. Since the legs were turned, and they tapered down toward the foot, it was almost impossible to lay the legs against a fence and hold them steady. There are a lot of ways to cut rabbets, including a hand-saw or hand-plane, neither of which I can manage well, so I finally decided to cut them on the table-saw with a regular blade, holding the leg upside down and vertical, and run them through the table-saw very carefully. I did have to make a small cut with a cross cut sled to finish the rabbet, but again, I was very careful.
I used table-top mounting clamps (Lee Valley #13K01.01) to attach the top, as they allow the top to freely move with the seasons. In order for them to work properly, a groove had to be cut on the inner surface of the long rails.
With the long rails cut to size (half of them with their ends mitred), I assembled two sub-assemblies, each of which consisted of two legs and two long rails. I glued the rails to the rabbets in the top of the legs, then added the wood pins. When they were dry, I flushed the pins to the rails surface. The inner long rail was fairly easy to position on the leg; its ends finished flush with the end face of the leg. The outer long rail was harder to position, but with some patience and planning everything worked out nicely.
With the two sub-assemblies lined up side-by-side, I attached the inner short rails to the rabbets in the legs with glue and screws. I then attached the outer short rails, which were glued and clamped over the inner short rails. Holes were then drilled through the inner and outer long rails, into the legs, and wood pegs were glued in place. Once dry, the pegs were cut flush.
At this point, all that was left was to get the table ready for finishing; all surfaces were sanded and all edges were eased.
Applying a finish
I’ve never finished a serious piece of furniture like this, so I sought the advice of a few woodworker friends. I got several suggestions but they all agreed on one thing: don’t use brush-on lacquer. So guess what I used? Brush-on lacquer. I found it easy to use, and I’m very happy with the results, but there are a couple of caveats.
First, lacquer stinks. I mean it stinks with a crash. And it’s as toxic as all get-out. If your eyeballs don’t fall out and your bones don’t turn to mush, then lacquer will just plain kill you, and the lacquer thinner is equally fun. So you need some serious ventilation, like a dedicated room or booth or hood and a respirator with cartridges rated for organic vapors. I finished my table outside. Finishing outside is not ideal – although I feel like it worked well for me – but you have to deal with little critters or debris getting stuck in your wet lacquer and the finish not drying evenly when in the sun. For me it was great. With a slight breeze, the odour and fumes dissipated quickly, and a little extra sanding removed any nibs and critters.
Second, you have to use a good real bristle brush to get the best results. I used a cheap brush and spent some annoying time removing the bristles that fell out of the brush and stuck in the wet lacquer. Basically, all you have to do is load your brush and make smooth passes in the direction of the grain, trying not to go back over the same spots more than once. Lacquer dries very quickly, and I put on six coats in two days. I sanded – just a quick pass with some sandpaper wrapped around a block of wood – between each coat to smooth out any little bubbles or nibs. I used a finer grit sandpaper with each coat. Luckily the lacquer odour is all gone in 24 to 48 hours.
I wanted to use a clear lacquer and get a real glassy finish, but someone I know (I won’t mention any names, but he works for this magazine and his first name starts with “R”) told me that I didn’t have the skills or experience to get a real glassy finish. So I used a semi-gloss, which probably hides my mistakes better than clear. All in all, I’m happy with it, and it’s definitely one of a kind.