Home Articles Forums Gallery Marketplace Videos

WoodCentral Forums

Knowledge You Can Build On — Est. 1998

ADVERTISE HERE - FREE TRIAL

Reverse Engineering a Piece of Chinese Furniture *LINK*

Posts

Reverse Engineering a Piece of Chinese Furniture *LINK*

#1

Reverse Engineering a Piece of Chinese Furniture *LINK*

charlie belden

I inherited a houseful of Chinese furniture, a few pieces from the 30's and 40s, but mainly pieces made in the early 70's, probably from Taiwan (though likely made in what was then "Red China" and therefore not importable to the U.S.). As a woodworker, I studied a piece - a sideboard - with drawers above and cabinet doors below (see attached link).

Thinking in terms of how "western" sideboards are done, I noted several significant differences

- no butt joints, mitered and triple mitered instead

- all the joinery is hidden

- illusions of light and relatively delicate ain't

What You THINK You See - Is NOT What is ACTUALLY THERE

what appeared to be several parts - cove molding on top of the apron

and a half bead molding under it - is in FACT all one piece of wood

AND

it ain't 3/4" thick, but more like 1 1/4" to 2" thick.

I did a cross section of the sideboard to illustrate the Illusion vs Reality.

Now I'm focusing on probably the most complicated joinery of that piece - the leg to apron joint - which is a triple miter - with a twist.


Working in Google's free SketchUp 3D modeling app, I created virtual, to scale, 3D models of the apron part and the leg part, then tried to overlap them as they would in use, and extract the simple part of the triple miter joints, the mitered surfaces on each part.


With the cove and beads details and the rounded corners, extracting the mitered surfaces was too confusing. So I simplified things a bit and started with simple rectangular parts, the leg 2x2 and the apron 1 3/4 x 1 1/4. Extracting that miter joint's surfaces was MUCH simpler, though not trivial since the front view of the miter is NOT a simple 45 degrees since the parts are not the same dimension.


Once I had that simple triple miter joint model, adding on the cove molding on top of the apron was relatively easy.


Adding the bead on the show faces of the leg and the underside of the apron and getting them mitered is going to take some thinking - and some trial and error. But what I did learn is that there was a lot more planning that went into this piece than simply cutting up some stock and routing details on a router table with fence.

I know how the mortise and tenons on the mitered parts are done. Modeling them to scale is going to take some work. Them Chinese furniture makers are GOOD.

Comments, questions, suggestions welcomed


Studying Chinese Sideboard

Re: Reverse Engineering a Piece of Chinese Furniture *LINK*

#2

Re: Reverse Engineering a Piece of Chinese Furnitu

Rick Haverstock

Charlie,

Nice discussion on the Asian joinery. You should get your hands on two books: Classic Joints with Power Tools by Yeung Chan and the Dover reprint on Classic Chinese Furniture (don't remember the exact title). Chan walks through the tenoned miter and the three way joint you illustrated. I used the miter joints on a pair of doors and they are slick and very strong. The Dover book has many measured line drawings and details of similar joinery for a variety of different furniture. I started playing with this after an article in Fine Woodworking on the kane tsugi joint (another type of tenoned miter) I lay no claim to expert status, although I have done some case pieces using the classic joining techniques. When I was stationed in Korea, I watched the local craftsman make beaded moulding freehand using a plane make out of a piece of broomstick with a beer can opener blade. Now that was a very humbling experience. Keep it up!

Rick Haverstock

Re: Reverse Engineering a Piece of Chinese Furniture *LINK*

#3

Re: Reverse Engineering a Piece of Chinese Furnitu

William Duffield, on the Cohansey

The title and author of the book that I think Rick is recommending is "Chinese Domestic Furniture", by Gustav Ecke. But, I can't imagine you have gotten this far without having and studying a copy of it, Charlie.

It is available from Tools for Working Wood.

http://www.toolsforworkingwood.com/store/item/AQ-1037

Re: Reverse Engineering a Piece of Chinese Furniture *LINK*

#4

Re: Reverse Engineering a Piece of Chinese Furnitu

charlie belden

I have both Ecke's books as well as Yueng Chan's, though I must admit I have only browsed through the latter. I was disappointed with the Ecke book. Linking the text description to the "plates" images and illustrations is, at best, a chore. Given that it was done in the late twenties or early thirties as I recall, it's not surprising that the photos are not very clear and the line drawings are OK though the dimensions and details leave much to be desired. I'm really surprised that there's so little information available on Chinese and Japanese furniture joinery. It may be that the western market for this information isn't very large and caters more to pieces an advanced amateur beginner or intermediate furniture maker might find they could do. Or perhaps it's because illustrating the details of a piece, and the joinery would be so difficult to do in a 2D medium. Were "publishers" to get into providing virtual 3D models that could be downloaded and explored, which they COULD do right now, maybe this Chinse / Japanese furniture joinery gap might close some.

If, or when, I get the virtual 3D 1 to 1 scale models worked out, I'll make them downloadable on my woodworking site, assuming the file sizes aren't HUGE. Will post my progress and provide a link to the SketchUp files when I'm satisfied with them.

Re: Reverse Engineering a Piece of Chinese Furniture *LINK*

#5

craigd

Re: Reverse Engineering a Piece of Chinese Furnitu *LINK*

craigd

For Japanese joinery, you might want to check out Chris Hall's series of books "The Art of Japanese Carpentry Drawings", Vols I-IV.


http://thecarpentryway.blogspot.com/

Re: Reverse Engineering a Piece of Chinese Furniture *LINK*

#6

Ellis Walentine

Speculation

Ellis Walentine

I suspect the miters are more complicated. Usually they were done with internal tenons. Here, since it is a case joint, there might be less need for additional reinforcement, but I'd err on the safe (and traditional) side.

On another note, I would make that vertical tenon larger in section, so that the mitered shoulders of the joint are only about 1/2" wide, depending on the scale of the piece. If you're clever, you can create stub tenons on the ends of your apron pieces that engage with grooves in the leg tenon to essentially lock the pieces in place. Full tenons would be better, especially if the joint isn't tied into rigid casework somehow.

Ellis

Re: Reverse Engineering a Piece of Chinese Furniture *LINK*

#7

Another book *LINK*

Keith Newton

Hey Charlie, Here is another book you might like, although it is more for architectural, some of the joints will make your head spin. I have to wonder how long it took those guys to make those joints.

I used to enjoy incorporating some of those joints in my work, but even when I would show my clients the complexity, they didn't get it. Patrons here are not as generous as they are in lots of other market areas.


http://www.amazon.com/Complete-Japanese-Joinery-Hideo-Sato/dp/0881791210

Re: Reverse Engineering a Piece of Chinese Furniture *LINK*

#8

Re: Another book

charlie belden

Craig & Keith:

Both sources appear to focus on relatively massive work and not so much on

furniture scale pieces and joinery. The larger the scale, the less significant

1/16" to 1/8" of slop becomes. At furniture scale, 1/16" of slop is iffy, 1/8"

is WAY sloppy - and compromises the structural integrity of the joint. If it

affects what's visible from the outside - open joints - especially triple mitered

joints - THAT is not acceptable.

There's also an issue of scale when it comes to strength of wood. What works

at a large scale may be too weak when scaled way down.

With all that said, both sources have interesting information that may be able

to be used at furniture scale, though I probably won't be spending $100 on the

four books on carpentry joints - even if it is Japanese ; ) I do have to say that

the Japanese temple builders kept both gravity (compressive loads) AND lateral

loads (earthquakes, typhoons and maybe tsunamis) in mind when deciding how

to build temples.

Re: Reverse Engineering a Piece of Chinese Furniture *LINK*

#9

Re: Speculation

charlie belden

Ellis:

Baby steps first - then work towards the 3 second 100 yard dash.

You are absolutely correct in guessing that the Leg would have a vertical tenon that goes through a mortise in the Aprons mortise. The Apron tenon into the Leg's mortise in the miter joint would resist vertical loads. But, sans side grain to side grain glue, does nothing to keep the joint from pulling apart horizontally, let alone resist racking. THAT is where the additional vertical tenon on the leg and vertical mortise in the Apron tenon comes in.

The context for the virtual 3D models of the triple miter joint is shown in the following illustration - sans the table legs. The table top is a frame and panel constructions but with the edges of the frame profiled with rounded outside corners and a bead at the bottom


The idea is that the the tall tenon on the inside of the legs would extend up through the apron and into the underside of the frame of the frame and panel top. The shorted tenon on the leg would go through a mortise in the tenon of the apron. So the apron would be pushed DOWN onto the short vertical tenon on the leg and into the mortise in the mitered leg top. This would prevent the apron from pulling horizontally away from the top of the leg.


The nice thing about these virtual 3D models of joints is that they're self documenting AND all dimensions can be measured. Because they're computer files, they're harder to lose and easier to share.

Re: Reverse Engineering a Piece of Chinese Furniture *LINK*

#10

Ellis Walentine

Re: Speculation

Ellis Walentine

I get it, Charlie. I guess you just have to figure out where you want to land on the continuum of safety here. If you anticipate horizontal forces that could pull the case apart, the added locking tenon/key becomes an elegant solution, though it complicates construction. We all strike a balance when we engineer joinery. I usually figure out the ideal and then back up to the appropriate.

Wish I had taken the time to learn SketchUp. Of course, then I'd be spending even more time at this computer. :/

Ellis

Re: Reverse Engineering a Piece of Chinese Furniture *LINK*

#11

Re: Speculation

charlie belden

When designing a piece, be it a table, as in this case, or a chair or a dresser, you have to consider ALL the forces the piece is likely to be subjected to. There are SIX degrees of movement at any joint, though we usually think of only three - In/Out, Up/Down, Left/Right. But - you can also have rotation about each axis - that can torque the joint.

The fun and games with tables - that have no lower stretchers - is quite different from case work. With case work you're dealing with basically a box with five of the six faces made of wood (unless it's IKEA) - one end open so you can get into it. And most case work isn't moved around very often.

Tables, on the other hand, get moved more often, usually lifted and moved or pushed or shoved. When lifted to move it, it's often from one end, two legs in the air, two legs on the floor. THAT puts a lot of torque on the apron/leg joints of the legs still on the floor. When a table is pushed, that puts a lot of torque on the apron/leg joint at each leg. If the joinery fails to account for the forces it will be subjected to - it will likely fail.

We tend to rely heavily on glue to deal with forces furniture is subjected to. The early Chinese and Japanese furniture makers had only "fish glue" and "rice glue", and their pieces were subjected to extreme temperature and humidity changes. So - they not only designed to accommodate environmental changes, but also relied as little as possible on glue - developing joinery to deal with forces mechanically. Both cultures also had over a thousand years of Joinery R&D and developed some very complex joints to deal with just about any situation. But the changes evolved slowly and adopted only after demonstrating, in use, that a new joint held up over time. It also helped to have an Imperial Family as your client / benefactor. That fashion trends changed very slowly also helped ; )

Sure wish someone would do an illustrated book on Chinese and Japanese furniture joinery. Yueng Chan has started on that herculean task. Having virtual 3D models of the joint in various stages of making would be REALLY nice.

Theoretically, once you have a virtual 3D model of a joint, you could run a stress analysis on it to find weak areas and over strong areas and come up with optimum joinery for a specific range of demands and with different woods. Through in fabrication constraints and it might be possible for an app to develop an optimum, frabricatable set of joints. . .