Category Archives: Timber Framing

Designing a Japanese Timber Frame



I’ve been asked where my daiku endeavours are headed after my study trip to Vermont. Seems like one thing keeps leading to the next, which is nice, and I’m always trying to learn new concepts and expand my knowledge. Among the odder things of late was being asked to interview for a Japanese reality TV show, “Who wants to go to Japan GP”. I’d love to travel to Japan one day, but it will be under my own terms, and this summer finds me far too interested in other things. I think I like a challenge, and buildings are often a series of problems with, at first, no perceivable solution.

Designing a Japanese timber frame is  daunting for the first time. I find myself staring off into the distance visualizing framing details in section view, re-reading for the twentieth time construction details in various books, and pacing back and forth trying to visualize what it would be like to walk through the structure.

To make it all come together requires a set of plans. I’ve abandoned the computer drafting and put together a drafting board and T-square, and borrowed my mother’s set of drafting instruments. (Ain’t it great to have a mom with shit like that? Love the woman.)


These are my text references. The two western books were my sources for understanding load/span calculations. “The complete Japanese Joinery” and Engle’s “Measure and Construction” are like two sides of a coin, old and new. Odate’s book is just fucking excellent (although its Engle who will show you fusuma, amado, and the all important tobukuro (amado closet).


For a frame of reference, this is where I’m starting from, a hole in the ground, half into decomposing granite bedrock. It will be a cellar, and I’m basically putting a little 10’x12′ shed over it.


The greenhouse was built right up next to the cellar hole, not enough room to get a small excavator in there anymore. It means taking material out by hand as the walls are chipped plumb in preparation for a re-bar reinforced concrete slip form masonry wall. I like to think I’m getting a little smarter as I get older, and put together a small counterbalance lifting device for getting the material out of the hole .

IMG_20160526_123421560_HDR (1)

It sits on a post with a level cut on the top, rotating on a pin made out of re-bar. The pin for the lifting arm is oak, it only took a couple of hours to put this together and has helped tremendously.


I have to deck over the whole cellar hole and provide enough depth to insulate the ceiling, so the total deck footprint will be about 15’x16′.  It doesn’t gracefully fit the ken building unit in traditional Japanese measure, which is about 6′ centers to the posts, but I think I’ve been able to come up with a balanced design.

In the above section view from the south elevation you can see that the big deck allows for a veranda, which will also wrap around to the north side.


There will be two sets of shoji, adjacent to the NW corner post.  Amado will cover the north and west veranda, an important element to keep from needing a drainage plane to protect the insulation infilling the floor joists. I also tried to design the roof to utilize available lengths and widths of metal roofing, but the effect of insulating the roof threw off my proportions on the west side and I had to extend the roof edge one foot to restore balance. I’ve been playing around with ideas for diagonal bracing too, I have to design for 130mph wind gusts and I don’t have certainty that nuki (half dovetail wedged horizontal bracing between posts) and good joinery will be enough to resist the shear forces imposed on the sides of the structure and ample roof margins.


I designed the floor system first, with much use of the cogged lap joint for the floor joists. The floor boards of the veranda run parallel to their length, and meet up partially with a mitered edge at the corner. This Japanese habit of stacking structural  floor elements on top of each other works to my advantage in getting enough depth to properly insulate the ceiling of the cellar. Although from what I can tell it is a departure to have the main floor joists in plane with the ground sill. The foundation will provide a knee wall to get the ground sill above grade in any case, so this stuff gets a bit fuzzy at times, you just have to make it work with what you need to build.


Insulating is another area that requires thought. I love the way the traditional clay infill looks, keeping the structural elements exposed is much less of a concern when you have an ample roof overhand protecting the building. That said, it gets pretty cold here in the winter, and if I want the structure to be habitable in the cold months a little insulation will go a long way. The rafters and decking of the roof is visible on the interior, a nice element I wasn’t willing to give up for thermal envelope, so it means insulating on top of the roof deck.  It creates a pleasingly thick roof edge, and even without a fascia board on the gable and eave ends covering the rafters can be used to create an interesting edge profile.


The gable roof framing is straight forward, the beam dimensions determined mostly by the loading and deflection requirements of the cantilevered overhang of the roof edges. Unfortunately I forgot to space some of the outer posts for the amado closets! On the west veranda it doesn’t change the loading much, but on the north edge I’ll have the noki geta beam carrying quite a bit more of the roof load from the ridge and purlins posts. This drafting work is time consuming, and I knew I’d probably have several design revisions.

Yesterday was frustrating in that regard. Okay, I see I need to move some posts, but where to put them. Okay, design the amado closet. Oh, wait, you need to know how thick your amado will be, how many, how wide. Oh, right, need to figure out a design for insulated amado that don’t look thick and ugly with insulation showing inside the veranda…


The walls present the greater insulation challenge. My posts are all 5″x5″ and don’t leave much room for insulation. I’m hoping that an EIFS system stucco with mechanical attached fiberglass mesh over poly-iso  will work.

There’s still much design work to come, for now I’m pleased to finally have some of my ideas to scale on paper, its a good start.