So, where were we?…
With the Indian Rosewood panel for the top 95% complete, I set to work on the frame pieces for the top. Here I was down to a reduced stock of Walnut, not having anticipated the need to make the top in the initial material purchase. I had enough to do it, but not the optimal choice of wood. A couple of the pieces had already been moving towards some other project in the shop and M had worked them over I guess with his orbital sander – I discovered the marks later. Fortunately there had been a change to that project and I was able to pry these pieces out of his grasp – it took a few passes with my plane to get those orbital scratch marks out of the wood.
The ideal choice would have been Walnut with very straight grain, in pure quarter-sawn or rift-sawn orientation. These pieces were a little less than perfect, but by carefully positioning the joint sections on straight grain runs, I could feel confident about about the solidity. You can’t just put a joint anywhere in a piece of wood. Driving a peg or wedge into a joined construction adds a loading force to a joint, compression or tension, depending, and this load will find any weakness. Add the loads that the structure may receive in use, and you can see the importance of considering the matter carefully. Again, coming from a timber framing background, so I think first and foremost about structure and joinery issues. Sloping grain, internal flaws, voids and splits in the material can lead to premature failure of a joint located in such an area, not to mention the increased difficulty this presents for cut-out.
Once I had the stock dimensioned a few hundredths of an inch over final size, I set to mark out and cut the joinery. Given the offset between the lower face and the upper face, so as to allow the panel to be as large as possible, the joinery was slightly more complicated than it would be otherwise. With a bit of time spent drawing, I was able to resolve these issues, making room for the panel to fit, and provide robust joint internals.
Once again, I choose the Japanese wedged locking miter joint, IMO the finest type of joint for this application, its only drawbacks being the time-consuming nature of its making, and the costs that associate to that. Among orthogonal joinery, this is one of the more complex. Why go to the trouble? Well, this joint allows for a miter on the visible faces, giving a pleasing continuity of grain around the joined corner, and the locking mechanism allows for good mechanical strength, without recourse to glue. I also has a discreet appearance, which I like. Of those who even notice this joint, in my experience, most tend to think it is some sort of splined or feathered construction, which it is not.
There are various versions of this type of Japanese woodworking joint – some wedged from the outside, some from the inside, some wedged at the corner, and some wedged adjacent to the corner (as in the top on the pentagon stool shown in an earlier post). In thicker pieces of stock, the internals of the joint are multiplied – I have seen tripled versions with 6 wedges on some applications. This version uses a single internal tenon and the wedges are doubled. The doubling helps to drive the joint together with a balance of force topside and on the lower side.
First off, I roughed out the joints, using a combination of table saw, hand saw and router:
Then I set about chopping out the slot mortises to receive their tenons:
Also visible in this photo are the stub tenons, or mechi:
It helps to have long paring chisels for tasks like these, this is a Tasai 10mm in the shinogi pattern:
Then I adjusted the mechi with my shoulder plane:
Here we are after a couple of pieces have been mortised. The two male ends visible have yet to have grooves cut into them to accept the mechi:
Here’s a male-female set, mechi complete:
Now I’ve processed the stepped rabate with dado for the panel on all 4 pieces:
Once together, check for square with each pair and adjust as necessary. These two are looking pretty good right off the bat:
After working my way around, I then clamp up all the pieces, and check it all for square and tight. Then a little fiddling as may be necessary, however there really isn’t much room for error. I leave the abutting surfaces just a hair fat so as to allow for some degree of room for paring-to-fit. One thing I learned in ju-jutsu is to pin the opponent with a light amount of force (this is the ideal, and means you have good technique) – in woodwork, I take this idea and apply it to clamping. Clamp lightly when checking for fit – if it requires thousands of pounds of force to squeeze together, the fit is poor.
Miraculously, everything clamped together and looking decent at this stage:
With the fit satisfactory, I could lay out the ramp-ways for the wedges. These one are offset halfway from each other:
And then knife the marked lines:
Here I use a small guide block and my ‘for kumiko’ dozuki saw, 32 t.p.i, to cut the ramp shoulders:
Then it’s time for a bit of chisel work to process the flanks of the ramp-ways:
Then I work on the more difficult female halves of each joint, this one is a little rough at this stage:
After all the joints are cut, I offer them up to see how the joints look:
This one is nearly fully-closed and is looking fine:
The next step was to profile the edge of the frame – I chose a semi-elliptical type of rounding the lower edge, and lightly chamfered the upper edge. The rounding was accomplished using some specialty Japanese chamfer planes I have in my set.
Here are the pieces of the top in preparation for final assembly. This follows a dry run assembly to be sure everything is fitting as appropriate:
Come together…
…right now…
…over me!
In go the wedges:
After the wedge ends are trimmed, it looks like this, with just a little planing required to clean the surfaces up:
And the miter looks like this:
It was the same result on all four corners, so it seemed redundant to post pictures of each corner.
The penultimate post in this series will deal with attaching the top to the frame, without metal fasteners, and a bunch of pictures of the completed piece. As you’ll see, Indian Rosewood has a deserved reputation for beauty.
Vanity IV is a great tutorial on the wedged mitred locking joint and a real eye opener for me! Amazing work Chris, thanks for sharing the construction details.>>Marv
Chris, can you tell me a bit about your kumiko dozuki saw? With its long tooth style at 32 tpi it must be a remarkable piece of work, the Japanese saws I have seen offered for sale seem to draw the line at 28 tpi. Is yours handmade?>>I would be really interested in reading some blog entries about the tools in your set. Probably a few interesting tales to be told there.
Hi Marv,>>actually that kumiko dozuki is Japanese, machine made, commercially available saw. The brand is Nakaya, D210-C. It’s Swedish Steel even. Contact Iida Kanamono, linked at the right side of the page, and see if they can get it for you. I suggest buying it with several extra blades, as the little tiny teeth are easy to break, even in soft wood. >>Thanks for the request about a posting on tools – I will make a plan then to talk about tools in upcoming posts. >>~Chris
Hi Chris,
I love this joinery. In fact it might just be my favorite… it's beautiful, light tight, and strong.
I've tried looking for more examples and illustrations online and in my joinery books, but have come up empty. Could you recommend a good source?
I'd really like to start practicing this one as it seems to be a very useful piece of joinery.
Thank you,
michael
Michael,
I don't know of any books that cover it in any detail, however I will be extensively detailing the joint in an upcoming volume of the “Art of Japanese Carpentry Drawing” series, probably vol. 5.
~Chris
Hi, I just want to thanks for sharing really useful article
Jolie,
thank you very much!
~C
too much info related to wood working… appreciated!