Half-Blind Dovetail With Mitered Shoulder: What Was I Smoking?

This weekend, I started work on a new project for the shop that will soon be desperately needed. As with nearly all of my other projects, I drew it up and decided on the joint–dovetails for a carcase frame, of course. I recalled that I’d seen a mitered-shouldered through dovetail joint, did a little bit of reading on it, and decided that it would be a neat one to try.

Except that I decided to make it a half-blind version. I clearly did not plan this one out very well, but I was very careful when laying out the tailboard, making sure to put the Xs in the waste parts, and indicating where the shoulder would miter. I thought that this would be a piece of cake, I’d just rough out the miter on both pieces when fitting the pinboard, then fine-tune it later. After all, both miters would be the same angle, right? Right?

What could possibly go wrong?

It should have occurred to me that I’d screwed up somewhere simply based on the fact that as I was sawing down the pinboard the first time, I managed to saw on the wrong side of one of the lines. At the time, I chalked this up to not having taken a break, the anticipation of the big game soon to start, and the fact that a really annoying song was playing on the radio when I made the mistake.

Unfortunately, it was actually fate trying to warn me, and I didn’t listen. Instead, I just sawed off the end of the pinboard, marked it out again, and went upstairs to watch the game.

When I came back the next day, I managed to saw everything correctly, popped out most of the waste, and then, to make sure that the pins and tails fit, sawed off a small amount of the miter on the shoulders.

The tails and pins fit without paring. That was as far as the good times went, though, because that’s when I got out my other T-bevel, set it to 45 degrees, and went to mark out the final miter for the corner.

Something wasn’t right. Why didn’t the miter line up to the corners of the joint? Everything should be the same width and thickness, ri–oh, wait. Duh, on a half-blind dovetail, unless the tailboard is thinner than the pinboard, the joint profile will not be square, and that’s not going to be a 45-degree miter. Furthermore, you can’t cut your miter beyond the half-blind portion if you’ve already done your tailboard, because the tailboard probably doesn’t extend that far.

I don’t know how long I stared at that thing, trying to figure out what to do. I didn’t know if I should try to salvage it, or just hang my head in shame and start anew with something a little more conventional.

Masochist that I am, I chose to try to salvage it because I hadn’t cut beyond the half-blind portion of the shoulder. For another equally long time, I tried to figure out what this was going to look like. I’m not sure I had an idea, but in any case, I started by marking out the miter on the pinboard from the pin base to the half-blind line (or whatever it’s called). Then I measured that angle: 50 degrees. That seemed really fishy to me, but I set my T-bevel to the 40 degrees necessary to complete the full 90 degrees and marked it lightly on the tailboard.

I banged the pieces together a little. Of course, I dented the beautifully-surfaced face that my newly-acquired Taiwanese plane had made because I used a buffer scrap that was too small. At this point, I didn’t care about that anymore, I just wanted those two miter lines to be perpendicular, and to my surprise, they were.

So I pulled the pieces apart and shaved down to the miter lines with my Veritas mini shoulder plane (this worked remarkably well). And then I banged the pieces together to see if they drew tight.

They didn’t, of course. In addition to this unsightly gap, there were also big gaps at the pinboard baseline, and I suspected that they might be related. I put the work down for the night and went off to freak out about something else.

When I dragged myself back the next morning, I tried jamming a piece of paper into the miter gap. It went in only halfway, so I pulled the pieces apart and checked the square of the long edge to the mitered surface. It was slightly out of square–kind of convex on the top. I pared it out (maybe making it a little convex in the process) and tried again.

I almost fainted. There’s no way that this should have drawn tight given the number of errors that I made.

The preceding photo was taken after I pared off the excess of the pinboard on the end–I wanted to see what it looked like. I’m afraid to admit this, but I sort of like it. Only I wish I could say that I’d actually planned it that way.

Now I have to make three more. Guh.

3 thoughts on “Half-Blind Dovetail With Mitered Shoulder: What Was I Smoking?

  1. I really like it. I have done lots of thru miter dovetails, but now I have to try your version. I think the key to the layout is that the thickness of the tail board has to be the same as the length of the tails themselves. This should make the miter right at 45 degrees. Your example looks super, imagine how cool that would be on the top of a drawer with a contrasting front.
    Good job,
    Gary

    • RB: Thanks!

      GB: Yes, to get a 45-degree miter, the thickness would need the tail height. The miter would still end at the same place, it’d just be easier to lay out. There’s a tradeoff there, because you’d have to be careful about the thickness.

      I thought about the “kerfing in” technique to get these matched without a T-bevel. I might do it on the next joint.

  2. I think the joint looks awfully good for your first attempt. I like how the tail/pin shows through. I wasn’t expecting to see this on a half blind mitered dovetail.
    rjb37

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>