Tapered Double Tenons
JMAwoodworks writes in, “Can you talk more about your tapered tenon universal joint, how you arrived at it, and broadly how you make both parts, and what application it might have in a home shop?”
That’s a big one, and I’m glad you wrote this in ahead of time, because I had time to think about it, and I almost decided to just dedicate an entire show to that, because it just does open a lot … And we may still do that another time. Double tenons have been around for centuries. The Japanese even have … I saw Toshio Odate one time demonstrating with a double chisel. It was a two-pronged chisel for chopping double tenons, so I know that the idea is not new. The idea of a tapered joint is not new, because Windsor chairmakers have been tapering the top of the legs as they go into the seats for centuries.
What is new, I think, I just haven’t seen anyone using tapered tenons like these, these two tenons, each, I don’t know if you can see it there, the tenons actually taper three degrees, so that they, over their 1 1/8″ … Excuse me, 1 1/4″ length, they taper 1/8″. That’s just done with standard mortising and tenoning techniques, but using a custom bit with a three-degree taper. Whether that’s practical in a shop has more to do with the attitude of the person running the shop. I would sooner have a tapered bit and a really good mortising and tenoning system and sell the table saw, tune up the bandsaw, and use bandsaw and router for getting things done. But the, after experimenting with this for years, the value of being able to efficiently cut a very precise tenon that comes together right off the machine is worth a lot.
Not only does it save time in being able to build really fine furniture, because, one, because you’re not having to hand shape or hand dress the piece, but the assembly doesn’t require touch-up. Things just come together perfectly. That’s … The ability to do that will drive your design thinking, and that’s even more important than the efficiency of getting the piece done, is … Our joint-making capabilities have been what drive our design concepts since we started thinking about making furniture. When I say “we,” I mean the human species, not just those who are paying attention to the show.
Practical in a shop, well, FMT jig, and other, the Multi-Router systems, all of those are capable of cutting an accurate tenon. Being able to cut an accurate tenon with a taper is just a matter of having a custom bit made. We can talk about that later, and I’ll probably post on chair … I will do that, I’ll make a commitment to post on chairmakers.com what the source that we get these bits from. They are expensive. If you’re going to buy one or two, I think they’re something like $250 apiece. If we get enough people wanting to do a group order, I can organize that. We need to order some as well, and they’re … They’re going to be a lot less, so I can, if there’s a bunch of you interested, send us a note, and we’ll follow up on that and see what we can work out. But I do think it’s totally practical in a home shop, but it means being willing to think a little bit differently and think about things that have a slightly more complex relationship.
What I have found is that woodworkers, for the most part, try very hard to avoid that, feeling that woodworking and chairmaking is complex enough without adding what seems like a complex joint. But once it’s mastered, employees of ours who haven’t been here very long are able to cut mortise-and-tenons like this on our system that are absolutely within tolerance. The other thing about the taper joint, and I’ve got another one here, this is one that we do on the outdoor chair, this one has a housed shoulder. In other words, the shoulder here is below the surface of the part. What that does is it allows us to have this curve be continuous without having that curve interrupted by a flat spot here, where the shoulder would ordinarily land. The whole piece goes sub-surface, so here’s the part. This leg was cut, we’ve been kicking this around for how many years in the shop?
Melanie: Seven or six. Six.
Brian Boggs: Six years ago, this was cut on our overarm router. This was cut last week, and there is a 10-thousandths taper on each tenon, so the tenon drops in nice and easily, and I can’t … Right here, I cannot push it all the way in. If I bear hug it, I might … No, it’s going take a clamp to finish that up. So these tenons are actually, when buried, slightly larger than the hole. Their seam is, I don’t know if you can see that very well, but that seam is absolutely airtight. These joints were cut a year apart, and I just went down-
Melanie: More than a year, more than a year.
Brian Boggs: Excuse me, seven years apart, and I just went down and grabbed these parts and put them together. Now, one thing that means is that when this was cut, it was probably the same moisture content that it is now, even though it has, certainly would have fluctuated up and down between now and then. But the benefit, one of the many benefits of being able to do a housed shoulder mortise-and-tenon so that the thickness of this part is the same as the total width of this joint, so this whole part is going into this leg, but it allows a seamless joint where you would otherwise have a problematic meeting of shoulders. Because the tenons are flush to the outside here, it means it’s like having this part with two legs instead of one. It’s going to be a lot more stable, a lot more resistance to torque and wrack. It’s about as strong a joint as you can get for these two pieces.
The other thing is, especially in this part of the joint, I don’t know if you can see that very well, but right here, there’s a little shoulder, a little shelf, essentially. This is where the front leg comes in. It’s exactly the same size joint, so this is going to fit in there just right. But because that shoulder … Now, this mortise is actually a three-way, compound, fully housed double taper locking tenon. That might not be practical in the home shop, because you can’t do that with a plunge router. I think you have to have an overarm router to do that, and the joint requires a tricky setup. But the advantage is that, with this shelf here, the weight of the person sitting is borne by the entire rail, not just its tenon, as typically they would be in most chairs, so there’s advantages there.
Another advantage in the taper locking joint is that it actually is more forgiving dimensionally. Now, these were basically a perfect fit. I don’t think we can do better than that. But these tenons can actually be three thousandths, maybe four thousandths larger than the hole itself, because they start off 10 thousand … Well, they would still start off six thousandths of an inch smaller than the hole, and they would be halfway in before they started getting very tight. With that much surface contact, they don’t tend to split the mortise. Plus, they’re already deep enough that they’re not going to shear all the glue down to the bottom of the hole as it would if it were a straight mortise-and-tenon. That’s even more true on this one, because this one drops into its mortise almost all the way home, just dropping in, and so it’s trapped all that glue between the tenon cheek and the mortise wall.
Now, there’s about a 1/16″ gap that you probably can see here. I can’t push that any deeper, but at a three-degree taper, a 1/16″ gap, means that the tenon is actually six thousandths larger than the mortise itself. But with that taper and the fact that it’s this deep in there before it gets tight, we can press this in here, and it will lock as it goes in. In other words, as soon as it bottoms out, you can’t take it out. Without glue, you can beat it out with a hammer, but you certainly can’t pull it out by hand. With glue, what happens is, it presses that glue to such a thin wall that it will seize up pretty much immediately. That means when you get it home, you can pretty much release the clamps within a couple of minutes. So there’s … It’s a tricky joint to learn and set up for, but once you get the hang of it, it’s an amazing joint to work with.
Here’s another thing about it. We do tests a lot with it, with these joints, and this was a chair that I didn’t like the way these shoulders came through here. This glue joint here was just, the glue line was really fat, and now that it’s seen several seasons-
Melanie: Show that glue line.
Brian Boggs: I don’t … Can you see that thick glue line there? It’s probably 10 thousandths of an inch thick, so when this came together, it didn’t come up like it should have. Now that I look at this, I can see that it did not bond. The glue isn’t going to bond if you don’t pull it up tighter than that, unless it’s epoxy. So we kicked this one, rejected this one, and decided to use it as a test, and so what we did is we put it in the bench vice and put one dog at this end and one dog at that end, and then just squeezed, and this leg flexed, or these two flexed closer together by 1 1/2″ before the leg finally broke. You can see that’s not short grain. That’s long grain. But this leg broke, the joint here never did move. Even after that test, and a few years later, it still looks fine. Other than the fact that the shoulder did not bond. I mean, that’s not a … You do get some strength from a shoulder seat, in a seal, in a bond there, and I want to get that, because the joint will look better over time, but-
Male: What kind of glue?
Brian Boggs: This was yellow glue, or PVA glue. We have used hide glue for that, but with this joint, there’s basically no advantage in hide glue. This is really not a fixable joint. I mean, if you’ve got hide glue bond in there and, for some reason, I don’t know how it would fail, but should it fail down the road, you’re not going to be able to get steam into that joint all the way around all of those surfaces and have it come loose, so … Hide glue’s just not worth the trouble, so we use the quick stuff. That was a long answer to a really good question. I appreciate the bringing that up, and we’ll get more into that another time.
