A Model Carpenter

As a carpenter, I can never really know what opportunities will come along in life, but there’s no denying that preparation is vital – that way, when I get the call to ‘come off the bench’, my tools and game will be sharp. One of the best ways of improving one’s ‘game’ in carpentry is through the study of layout, and that is well-expressed by making models to explore different challenges. Layout issues in roof and stair work are numerous, and can become quite complex. While solving these problems with 2×4’s is one level of difficulty, trying to do complex non-orthogonal connections with timber using joinery is considerably more difficult. Thus the model should be of a scale that permits realistic joinery to be cut where possible. I like to work at 1/2 scale, or for larger structures, 1/10 scale.

Model-making has a strong tradition in carpentry and woodwork. Shipwrights make half-section hull models, stair-builders build stair models. In Japan, as part of the study in preparing to build large and complex temple structures, it is not uncommon to execute a 1/10 scale model, some of which can take up to a year to make and cost as much as an average Japanese house! This may sound like a frivolous exercise to some, however the model-making allows for last minute design refinements, and less potential waste or design mistakes. And, in the case of Japan, bombed to shreds in the second World War, in some cases the temples are gone and all that remains are the models – thus they are important historical artifacts. Here’s one such model:

In Japanese carpentry study, there are in fact 2 formal exam levels: level 2 (the ‘easier’) and level 1 (more difficult). The former level 2 exam, investigates the splay legged form, a 4-sided prism, is identical to that of a steeply pitched roof. Here’s what that sort of sawhorse looks like:

Upon first glance this may appear to be a simple structure, however there are many subtleties – the simplicity is deceptive. The legs, for instance, aren’t square in section:

A square-section leg, when splayed on an angle in two directions, will not form a square section where it meets the floor, or any horizontal plane- it will be diamond-shaped. The Japanese approach is to reshape the leg into a diamond shape so that when it is splayed it meets the floor as a square section. On top of this the stretchers are tenoned right through the legs – each half-tenon just touching the adjacent one on an arris.

Anyhow, the exam gives the carpenter 6.5 hours to draw the model, complete with developed geometric views of the leg and mortise layout. Once the drawings are complete, the student takes material that is 1 mm oversize and then hand-planes the pieces, forming the diamond-shaped legs, and the other parts. This is an extremely rigorous test, primarily because of the limited time frame. The drawing alone takes close to an hour. The Japanese greatly value both skill and speed in carpentry work.

I can make one of these sawhorses in about 8 hours, but to get it down to exam speed I would need to train.

The sawhorse, as I mentioned, was the old exam subject – the new exam is essentially the same, except the sawhorse is now, as it were, inserted into the roof, on it’s side:

This model I did while teaching timber framing at the College of the Rockies in B.C. a few years back. Note that both askew common rafters flanking the central common are parallelogram-shaped in section. this enables the rafters to form a flat section on top, in plane with the roof, and meet the wall plate evenly. Again, a 6.5 hour exam covers this, including the developed drawing.

The level 1 exam takes the difficulty up a notch, and concerns the construction of a hip corner assembly with an askew jack rafter:

Again, 6.5 hours is allowed to draw and make this, all with hand tools. The first time I made this model it took me three days, and I am having trouble figuring out how to do the drawings, more complex than for the sawhorse, in under an hour. Again, I need to do some serious training! This model may look basic enough, and the difficult bit really is only that one ‘funny’ jack rafter, however do not be deceived by the apparent simplicity.

Here’s a variation on that model:

The ability to form parts, such as rafters, that can be put at odd skew angles and still meet the support surface and roof plane cleanly has many applications. Here’s one where the cantilever beams are skewed:

Another example of using rafters in this manner:

Level 1 may be the hardest level in Japanese carpentry, and there are only about 100 carpenters in Japan who are certified at that level, but it is not the most difficult level in Japanese carpentry – only the most difficult that fits the test format. Adding curves, for instance, adds considerable difficulty, as does working with irregularly-sloped intersecting roof planes.

One area I have devoted a bit of study to in the past is the reciprocal roof form, or, as it is termed in German, “Mandala-dach“. Here’s a model I completed while working in California:

This roof structural form is rather rare generally-speaking, however the greatest development of their use, and their refinement is seen, no surprise, in Japan:

These are some more pictures of the Bunraku Puppet Theatre:

And the exposed roof work from the inside of the building:

Sometimes I teach courses in Japanese compound joinery, and the subject of that course is a 1/2 scale hip corner model that looks, partially-assembled, like this:

The course to make the hip quarter section takes a full 5 days to teach, and involves some relatively complex joinery, especially where the wall plates (termed keta in Japanese) cross and carry the hip rafter, and a fair bit of math. It is one of the ironies of life for me to find myself teaching mathematics to others, given that it wasn’t exactly my favorite school subject!

One development from the straight hip rafter roof is the curved hip roof. This is a model I made while working in California, done during the evenings and weekends during the time I worked on Client E’s place:

I’m greatly looking forward to the opportunity of doing such a roof, in full, for a client. The above roof corner, I should mention, is displaying half of the actual set-up. This is only the decorative roof portion.

In some parts of Japan carpenters get together and form study groups to explore carpentry through roof models. Pictured below is an irimoya roof model:

I’ve begun the same thing here in the North East. I set a couple of prerequisites for potential study group participants, the second of which is the sawhorse pictured at the top of this posting, and after 4 months now only one intending participant has completed the two projects, so not much is happening with the study group as of yet. I might have to abandon the idea and revert to offering courses, we’ll see. I am wondering if an organized course format might suit most people better than setting self-study projects.

In Japan there are also regular carpentry meets, daiku gijutsu tai-kai, where special roof model joinery challenges are set and participants race to complete them. Here’s a view of such a scene from, the 43rd annual meet:

And another shot showing a participant with his finished model:

It would be great to do this sort of thing in N. America, however the problem is that there are so few carpenters in his country, as of present, who could actually make the models at all, let alone in a tight time frame. Perhaps though, the dangling carrot of competition would motivate?

And finally, in France, when a campagnon has completed his apprenticeship, it is normal to make a ‘Masterpiece’ (termed chef d’œuvre) where they can demonstrate their acquired skill and technical virtuosity. Here is one such piece, a photo sent to me by a fellow timber framer who visited France:

This is similar to the Trépied Établi I am currently studying – actually, that item is visible in the background of the photo, to the right.

And, for the pièce de résistance, look at this masterpiece, completed in 1955 by Ephrem Longépé:

This is quite a piece: splayed legs, cross-braced in every orientation, elliptical top frame and fitted reciprocal floor, elliptical dome with both round turret and ‘Imperial’ turret, and with a dormer on a bias. Magnificent! This shows a carpenter who can connect any piece of wood to any other piece of wood, at any rotation, angle or curvature, and do it with elegance and clean joinery. This is accomplished carpentry my friends.

Post II comes next in this series.

3 Replies to “A Model Carpenter”

  1. I would love nothing more, if you are still interested in doing a focus group that is, to participate. Now you know that I have close to no woodworking skills yet and I am not sure that the prospect of exchanging thoughts and experience with someone like me would be valuable to you but if you see any value, I would be interested in hearing from you and would be determined to assist in any way I am capable. I am very curious about the French saw horse (dah) though it might be way out of my league. I do need a pair of saw horse though lol. Excellent content as usual Chris, really excellent blog, I am looking forward to my daily read.

  2. Hi Chris

    Really interesting topic.
    I think that making prequisites is a good idea in order to keep the standard high, so you wouldn’t feel like you would “babysit” a group but that it would be a genuine chance of some knowledge exchange.

    The slow response from the people interested might be due to other pressing projects, or that the tests required are more difficult than what they might initially have believed that they could easily accomplish.
    I know that I would feel embarrassed to call back and tell someone that: “Sorry, but I am not as good as I thought I was. I can’t make these test pieces at all.” It might be easier to just not get back in touch.

    Not all carpenters think that traditional joinery is interesting.
    I had a carpenters apprentice helping me out almost ten years ago while I was adding a new roof on the barn. The steel roof plates were some 26′ long, so I needed someone to help negotiating them into place.
    At a time when doing the sloped gables, I told him that I had often thought about building a small barn using classic timberframe joints. His reply was: WHY??
    I was dumbstruck- His next line was, What is the point when using steel brackets, screws and a nail gun is much faster.
    I felt sorry for someone learning a trade and not in the most remote way being interested in the “history” of his own trade.

    I hope that you manages to find enough participants to make a study group.


    1. Well, thanks so much for sharing.
      I did run a study group for several years, however over time the number of folks who were able to work through the exercises and complete projects dwindled and dwindled until there was only one or two members, and then finally one. We pushed on for another year with one member only but eventually it all ceased. i think self-study exercises online are difficult for a lot of folks to get into, and many people have very busy lives, so making time for something difficult gets harder to do over time. Also, there are folks very interested in the drawings aspects but without the means or experience to tackle the woodworking part, and I insisted that the two had to go together, as one informs the other.

Anything to add?

error: Content is protected !!
%d bloggers like this: