It's not perfect. Looking at it, preparing to write about it, I see flaws all over. But, I've finally cut a dovetail I don't consider horrible.
For years this joint has eluded me. Always too tight off the saw, but filled with giant gaps once together. Results were no better with a router - in fact, I ruined a bit and took a nice chunk out of the jig while trying.
So what changed? Some of the expected: I have more practice, in general, and I treated myself to a few new tools. But, there are two elements that I think played roles at least as important.
Improvement number one came from Matt Estlea's videos. My skill in following written directions is second to none. I've read several step-by-steps, and attempted to perform their processes faithfully. Yet somehow watching Matt do it, listening to him talk through all the things he's thinking about as he's thinking about them, just made some of the important steps click as I was doing it myself this time.
As just one specific example, I've never particularly liked sharpening. I've read how to do it properly. I've bought nice stones to do it with. My edges have turned out well when I've done it. But it has always felt like a major chore. Watching Matt explain the reason behind the two bevels, and then show that it really is just a few strokes on the finer stones to keep the cutting edge in good shape made it a super easy thing to do, as often as I wanted. Quick, easy sharpening meant I (almost) never wasted my energy forcing a dull chisel to do awful work.
The second major element in my improvement came from experimenting, and closer evaluation of each result. This project is a box, and thus called for four dovetails. For the first, I used my new tools, thought about what Matt had demonstrated, and was close right off the saw. After that, I fixed the joint in my usual way, but ended up sort of "meh" - not my worst, but not enough better to make me excited about doing more.
I analyzed what had gone wrong. The joint didn't fit to start; the pins just wouldn't go in. So, I trimmed the pins to make them fit. By the time I got them pushed through, though, there were gaps around their visible ends. There are two simple explanations for this problem: either I pared a taper into the pins (so their tips were smaller than their bases), or there was already a taper on the tails (so the holes for the pins are larger on the outside than the inside).
Either problem indicates an error in sawing. I need to fix that, but in an attempt to get a feel for my new saw, I had decided to do all of my sawing up front. I needed to find a better way to fit whatever I already had. So, what about trimming the place that won't show instead? Bingo: shaving the inside of the tail instead of the outside of the pin produced a much tighter dovetail.
My final problem was how to move more quickly. Dovetails one and two each took an hour and a half or more. If I had sawn true, things would have been significantly quicker, but test fitting went slowly. It was hard to know where to trim. This is when I pulled another YouTube tip out of my history. Larry Potterfield perfects ill-fitting pieces all the time, by coating one piece in some sort of carbon black, fitting it to the other, and then taking them apart again to see where they touched.
Obviously I didn't want to coat the entire end of my board in carbon, but just a few scribbles of pencil on the cheeks of the pins would never been seen. Each test fit left smudges on the tails exactly where the pieces touched, darker where they squeezed harder. This cut down the guesswork, both speeding up the process and making sure I wasn't weakening the joint by removing wood from the wrong places.
Joint number three was the first dovetail I've cut that I'm actually not completely unhappy with. Dovetail four was almost as good. A little overconfidence, or a little eagerness to get onto the next part of the project, may have played a role in some small mistakes. I'll know soon: the dovetails on this project are practice for a much bigger project on the horizon.
Categories: Woodworking
Post Copyright © 2019 Bryan Fink