Desk No. 3 – Design 2

This past week as I’ve moved forward with this desk project, including buying and sorting some lumber and doing a bit of rough sawing, I’ve also been tweaking the design a bit here and there. The understructure continues to be a tricky area to wrap my head around. Among other things, deciding on joinery is always a balance between the strength of the joinery and it’s effectiveness, along with the ease of cutout. On top of that are considerations like demountability/repairability.

For this project I had initially thought it would be interesting to design this desk to be completely demountable, but little by little that plan has changed. I am still committed to using solid wood joinery throughout, but I’ve given up on the idea of being able to completely break this desk down to it’s individual parts. The legs will still be easily removeable by using wedged kata-sage-ari joints, but I think desktop understructure will end up being more or less permanently assembled. The only joints I could think of using to make the understructure demountable were sliding dovetails, but as I mentioned last time those joints also have the negative effect of weakening the parts that receive the dovetails. So I’m shifting towards using wedged through tenons to join the individual parts of the understructure.

The desk is long and I want to give the top as much support as possible using the two long apron pieces seen below. The taller I can make those pieces the more support the desktop will have, but of course you also have to consider how you will sit at the desk and the available leg space you will have. So I think I’ve come to a decent compromise regarding the size of these long aprons. But as I mentioned in the previous post joining the intermediate cross pieces to the long aprons using sliding dovetails will effectively make them much weaker; like cutting a chunk out of a beam. So I decided to go with wedged through tenons for those connections.

And I started to wonder, if I’m going to use through tenons to join those parts, why not for joining the long aprons and end cross beams as well? I started to entertain the idea and modified my model a bit to see what that would look like.

Overall I think using wedged through tenons in this area keeps the structure unified to the extent that all these joints on the understructure now have the same joint and appearance. I was also not super keen on having one sliding dovetail pass through another as would have been the case with my previous design (below).

Cutting a sliding dovetail through an adjacent sliding dovetail like this introduces weak areas where the end of the long sliding dovetail could split off, rendering it pretty useless. So there’s another win for using through tenons in my opinion.

Now on to something different. I’ve been contemplating how to join the desktop to the understructure and the idea of using sliding dovetail keys popped into my mind. I used this technique on a table I built back in 2015. It was one of the first projects I completed after leaving art school, and I was desperate to move away from the foggy abstraction that was so pervasive in school, towards something really technical and craft oriented. The result was this 5 legged coffee table with a pretty elaborate structure. I used solid joinery throughout and wanted to follow that trend through to joining the top as well. With a solid round top that had an exposed edge all the way around, standard sliding dovetails were out. So I went with sliding dovetail keys, which have the benefit of securing the table top while being completely invisible once assembled, and they still allow for wood movement.

Cutting these joints involves cutting a dovetail slot twice as long as the desired key and widening half of it to allow the key to be pushed down into the mortise and then slid into the dovetail slot. In the picture below I’ve got one key half-way slid into place. This was the first time I used these joints, and I’d do a few things differently now, including reorienting the grain of the keys to run perpendicular to the joining members for a stronger grain orientation. But that said after 8 years or so this table has been holding strong.

One tricky aspect to using these joints is the layout. For this table, after cutting and installing the inserts into the understructure, I flipped the whole assembly over and marked the actual location of each joint directly onto the tabletop itself.

From there I used a set of mdf templates to cut out the mating mortises and dovetails into the underside of the top.

Here’s the finished top with all the joints cut.

Now to bring this all back to the current desk project…. The stop messing around, get it done, practical side of me still thinks the desk would be pretty rock solid if it was locked together with a set of sliding dovetails on each of the wide end cross beams, along with some screws in slotted holes to add some reinforcement through the intermediate cross members. But the woodworking fanatic part of me really wants to go all out and use joinery instead of screws for the intermediate cross members. …And I think that’s where things are heading.

I’ve been scratching my head all week thinking about how to implement these joints, and I think I’ve finally started to hone in on a final decision. As always there are a few trade-offs and design changes that I needed to make. So next week I’ll cover this last bit of joinery in more detail, and here’s a sneak peek of things to come.

Thanks for reading.