Screen Play and then Benched

For the past few days I’ve cut hewn no wood and, for that matter, drawn no water. I’ve been designing stuff, in my head and on the computer.

On one front, if you’ll excuse a pun, I’m drawing a tsuitate, which is a Japanese freestanding screen, following a suggestion from a friend. This is a piece I am going to build and put in a gallery.

Here’s a few examples from Nippon, the first being what you might call a more-or-less typical version with grill work:

Just like shōji and ranma, the potential for variations within the form are nearly endless- a tsuitate is pretty much a blank canvas in a frame:

Nice panel, but I find the lower portion clunky and the feet are too small – and I don’t like that they are not centered with the frame verticals.

And for those proficient in their work with kumiko, the options run from mild to wild:

Amazing what you can do with a hexagon, isn’t it? Too bad the piece has such clunky feet, at least for my taste.

I’ve also moved along the design for a workbench some few hours. I thought I’d share with y’all how that is shaping up at this point. I’ve moved away from playing around with complex Mazerolle-esque bracing arrangements, and found that if you want to have drawers, those angled braces make that quite problematic. The splayed form is inherently quite stable and strong, so the braces aren’t really needed I don’t think. This piece is going to be a tour de force for splayed leg work, Japanese style.

My inspiration comes from three places. First, the disturbed and chaotic locale known as my mind, and my own odd ways of going about solving a design challenge, namely to make a bench that resists racking through structure and not by reliance upon metal rods and/or bolts, and is well sealed so that dust can’t get into the interior. And a bench with a pleasing form too! I went away from a slab top and instead opted for a rather heavy duty version of frame and panel, as you’ll see soon enough.

The other inspirations were a bit diverse. One was the following piece, one of my favorites from Koizumi’s fine book Japanese Furniture (originally published as Wakagu):

That piece may look simple, but I can assure you that done properly with backed posts and mortise and tenon joinery, it would be a time-consuming challenge.

The other source of inspiration is from a structure like this, a kake-zukuri form of Temple. ‘Kake‘, 懸け, means ‘hanging’, and -zukuri means made/constructed – these structures are perched on hillsides and require extensive scaffolding – this one is called Kongo-in, one of my favorites:

Put those ideas together, and what my warped mind comes up with is something like this (elevation view of the frame on the long side):

The outside posts are regular slope, regular plan compound splay. The central posts are simple one-way splay, which are easy-peasy to layout and cut. The middle posts are the toughies, as they are irregular slope, regular plan compound sloped legs. I’ve only tackled that sort of problem once before, in the irregular slope sawhorse that I continue to use as my workbench. That build was detailed last year, and can be found in the archive to the right of the page. It is a challenge I look forward to re-visiting, as I’m fuzzy on some of the details right now.

Here’s a perspective view of the main frame components:

The bench is about 90″ long and 30″ wide. The table top frame is 3″ thick and the legs are about 2.5″ thick. All the legs are connected by penetrating stretchers, or nuki, and in order to fit those cleanly all the compound sloped legs need to be backed on their faces. With the cross pieces in the frame through-tenoned in the legs as well, all four faces of the compound sloped legs will be backed. That’s the genius of Japanese carpentry my friends – while it is a tricky process at the outset, backing all the legs makes the joinery a snap, relatively-speaking.

Unlike a standard splayed piece with nuki, since I also have the requirement to fit drawers, I will make the nuki parallelogram-shaped in cross-section so that their top and bottom edges are in plane with the floor, which will in turn allow for drawers to be fit fairly readily:

The top panel is 1″ thick and the battens that support it and tie the table top frame together, will also be fixed to the underside of the top with sliding dovetails. The battens are fairly stout themselves, so I can place bench-dog holes through the top and into battens at their locations if I like.

I am wanting to make the drawers so that they can be slid out either side – that is, a one-piece drawer to span the width of the bench instead of a short drawer in each side. Or maybe some combination, with one-piece drawers in the top two levels and separate drawers along the bottom tier.

I’ll use some wrought iron Japanese drawer pulls, though I haven’t decided on the pattern yet. I haven’t decided upon the drawer slide system yet either – always a tricky point. Metal slides are indeed practical, but I always feel like making some sort of wooden slide, probably using lignum vitae….

I’ve detailed the top frame so that there will be no avenue for dust to get into the interior spaces and drawers:

I guess I’ll need to sort out some sort of dust seal for the drawers to their surrounding frames as well.

I figure I can make one end of the bench a place to hang a few planes, and I’m currently percolating a few different ideas through the devastated wasteland known as my cranium for various vice/hold-down and planing support fixtures. It won’t be too fancy in that regard, after all I’ve been managing okay with my sawhorse and Bessey clamps for the past while.

I won’t be building this bench for a while yet, so these are indeed early days in the design process. I’m liking the way it is looking at this point, but I may have it all wrong, who knows?

The tsuitate build however will proceed. The first post in that series is found here.

15 Replies to “Screen Play and then Benched”

  1. That's going to be one very cool looking and stable bench. Contents of the drawers may interfere with use of the bench dogs, however, especially if you are going to use hold downs.

    On question – what does “backed” mean, as in “all four faces of the compound sloped legs will be backed”?

  2. Hi Chris, I've recently just caught up having read all of your posts from the start. I am really loving this blog.

    Few comments that are intended to assist.

    On the Battens that run the depth of your bench, they sit proud. Will they interfere with say, a Bessey clamp reaching its full throat depth? Same question with regard to the Nuki/stretchers.

    Have you read “Workbenches From design & theory to construction & use” by Christopher Schwarz? It's a great read in my opinion. The best book on benches I have yet to find.

    The important points, in my opinion:
    1. Can you clamp a door to your bench so as to plane the face, edge and end?
    2. Is it of sufficient weight to resist racking/movement?
    3. Will it easily complete/assist with any task you will be required to do in order to complete a project?

    It recommends and I see the benefit in writing a list of tasks you frequently do, and seeing what features are required to do such a task. Then implement those features into the bench.

    In the drawings, it appears you have battens evenly spaced to as to create 5 equal sections. Assuming this is correct, 90″/5=18 inches between bench dogs. Have you considered the likely end vice will not have an extension of 18 inches (450mm)? Yes, you can use blocks to wedge a piece in. Do you wish to find a block when an unusual piece arises? is it not easier to place more dog holes closer together?

    I personally enjoy sitting at my bench occasionally. Your drawers would make this impossible. Do you like to sit at a bench while you work?

    In the sketchup drawings, you seem to have included a bottom plate which I assume sits directly on the ground. Have you considered the situation in which you meet a floor which is not level or which is not flat? I prefer the feet meet the floor myself. Or failing that, if a base plate(?) is used, then the recess of material to as to make small feet which sit proud of the plate. Hench feet on a plate. This still provides the bracing I infer you are going for, but also minimises issues where the floor isn't flat.

    I immensely enjoyed reading each of your posts and it has spurred an interest in international architecture/carpentry I previously lacked. I have begun to read books about this and am really enjoying myself. Thanks for inspiring that.

    And a question or two, could you explain why the splayed leg structure is good? I am en engineering student who has four years drafting experience at engineering firms etc so this is of interest of me, but I'm confused. The bracing in the structure is what minimises racking? The splayed legs widen the centre of gravity making it more difficult to tip the sawhorse/etc? But doe the movement of the bench not rely on the weight of the sawhorse/bench itself and the frictional forces applied by the flooring surface? If you could shed a little light, It would be greatly appreciated. What benefit do splayed legs provide?

    Tom Halliday, Brisbane Australia.

  3. Hi Chris,

    Nice looking bench. A great idea to have all the underneath divided up for drawers and useable space (not just a token amount).

    One of the faults I find with contemporary bench design (even if based on older designs) is that they are far too massive. It seems people expect to be able to jack-hammer on them or split firewood. Also, these benches are so large and heavy as to be rendered immovable and thus stuck in the corner they are initially placed in. Your bench design seems a reasonable compromise between stability, functionality and I suspect, portability.


  4. The splayed of the legs seems to extend beyond the edge of the top. With that and not toe kick space, you would not be able to stand straight with your body against the bench. Which seems to be how I chisel sometimes. Especially toward the end of the day….
    Maybe have bottom chords so only the four corners make contact with the floor, easier to shim the feet. You will always have to shim the feet…
    Good point above about bench dogs poking down into the top drawers. This can be a problem close up to vises as the dogs hit the hardware below.
    I personally like too think I could split wood on mine ;^)
    C on da BIG

  5. Hi Chris,
    just wrote a comment only to have it disappear… Have enjoyed your blog over the last several weeks and look forward to where this bench project is going. I agree with Tom's comments, highly recommend Chris Schwartz's book for getting the creative juices flowing and focusing on the important features of a bench. My imagination suggests that to sit at your design would be difficult and the lack of toe space annoying. It obviously a very stable design, but will it compromise the ease of holding work?
    good luck with it.

  6. Well, some great input there gentlemen! Who woulda thunk so many had so much to say about workbench design?! That's nice to see, and much food for thought.

    @Mark; who wrote, “One question – what does “backed” mean, as in “all four faces of the compound sloped legs will be backed”?”

    That is something I will make the subject of my next post, and I wonder how many other readers might have been wondering what I was talking about?

    Good point about the potential for interference with long bench dogs and the upper drawers – I hadn't thought about that and it is the perfect post-design and build WTF?! moment I was hoping to avoid….

    @Thomas Halliday; thanks for your extended comment. I haven't read or come across Schwarz's book, but I'll try to get my hands on a copy.

    You wrote,

    “The important points, in my opinion:
    1. Can you clamp a door to your bench so as to plane the face, edge and end?
    2. Is it of sufficient weight to resist racking/movement?
    3. Will it easily complete/assist with any task you will be required to do in order to complete a project?”

    I do believe my bench as it stands could configure the door so it could be worked on in all those orientations, though I could widen the top a little more I suppose to accommodate a full-width door.

    I fail however to see how the weight of the bench does anything to resist racking, rather I would say that excess weight of most bench designs, along with the typical plumb legs are factors contributing to an ease of racking. Instead of using the geometry of the parts to resist racking, most bench designs rely upon large sections of wood along with metal bolts and rods to keep things together over time.

    Now, a heavy bench would tend to resist sliding around on the floor when pushed sideways, however when my bench, as currently designed, has it's drawers full of stuff, in addition to the weight of the bench components, I think it will be plenty heavy in that regards. Also, I intend to put rubberized leveler feet in under the floor sill, which will further resist any tendency for the bench to slide about.

    As for point #3, I'll think about that some more.

    As for sitting at the bench, sometimes I might like to do that as well. The drawers, as I mentioned, will slide both ways, so they could be pushed out to one side to clear my knees and toes I would think. That is a good point to consider though, and I am now thinking that if I widen the top by 6″ each side, that problem would be taken care of – but I'll see if I can confirm that by drawing first.

    Finally, the sill on my bench will be hollowed slightly between post placements, and the portions of the sill directly under each leg will have leveler feet installed, so dealing with an uneven floor is already-thought-about item in the design.


  7. @Steve, thanks for your observations.

    I too have no plan to split firewood upon the bench or pound on stuff atop the bench with a sledge hammer.

    My issue with massive slab tops is their propensity for wood movement and warping over time, and the need for re-flattening. I don't tend to like slab construction of any kind and invariably opt for frame and panel, though in this case with a 1″ thick panel I may be pushing things a bit.

    If you look at 400 hundred+ year-old Ming Chinese furniture you will not see slab pieces, other than one form of table which had a thick fully dressed plank running between two smaller table supports. I also have noticed, having visited many East Coast Museums, observing many pieces of early American furniture, that slab table leaves and slab chest lids, etc., are very often quite warped after many years have gone by. I observe that the frame and panel pieces do not exhibit this tendency. This is one of the reasons that I stay away from the whole Nakashima-imitation industry that has sprung up of late making slab table and bench pieces. Quick and easy to do, especially with a wide belt sander, and the market wants to see 'real' wood so it sells I'm sure, but I think in the long run such pieces will have issues.

    @C on Da Big; there must be some optical illusion going on, because the top and the sill in the current drawing are exactly the same size. Toe kick room though is worth thinking about, and in concert with previous suggestions made above, I am looking to widen the top about 6″ each side. The other issues you mentioned I have already addressed.

    @Michael; well, I intend to design the piece to satisfy my anticipated work needs, however I have been able to find ways to use a sawhorse for most everything so far, so I can't help but think that the bench will add a lot of capability in regards to fixing the work in place. I've never found that issue to be a big deal – for instance when working with a pull plane, all you need is a catch at one end to pull the board against – no need to lock the piece down like as is common with Western-style planing practice.


  8. Also,

    Tom Halliday wrote,

    “And a question or two, could you explain why the splayed leg structure is good? I am engineering student who has four years drafting experience at engineering firms etc so this is of interest of me, but I'm confused. The bracing in the structure is what minimises racking? The splayed legs widen the centre of gravity making it more difficult to tip the sawhorse/etc? But does the movement of the bench not rely on the weight of the sawhorse/bench itself and the frictional forces applied by the flooring surface? If you could shed a little light, It would be greatly appreciated. What benefit do splayed legs provide?”

    Good questions.

    While the splaying of the legs does widen the center of gravity and this helps resist overturning, it is the splayed form itself that resists racking by the mean in which loads are transmitted through such a structure. By 'racking', I mean a lateral (horizontal) load imposed on the bench top, say by a large timber placed on top of the sawhorse and moved sideways/lengthways. The splayed leg directs some portion of the horizontal force down to the feet, that is, it transfers load along the grain of the leg. trees are 'engineered to be strong in that direction.

    A plumb leg, on the other hand, has only the joinery connection to resist the racking force, which invariably means that the load is converted into crushing the side or face grain of the tenon in the mortise.

    It's easy to demonstrate the difference for yourself. Stand with your feet shoulder-width apart (i.e., with your legs 'plumb') and, trying to keep your body straight, have someone push you sideways at the shoulder. You will tip over quite easily and it is physically difficult to resist the load. Your skeleton is simply not aligned to help out. Now put your feet 3~4 feet apart and have your friend try pushing your shoulder again – you will find, especially if you keep your legs locked straight (this is analogous to transferring the loud 'along the grain' – that is, along your skeletal alignment) that it is much easier to resist the push, simply by your body geometry. There is a certain point, with your legs about 45˚ splayed, where your ability to resist is greatest. If you keep your knees somewhat bent, you will still be able to resist the load, but it will require you employ muscle power in addition. This would be akin to using splayed arched legs in a constructed piece, where the load pathway is not directly along the grain but rather induces the leg to flex.

    A final point about the splayed form is that when racking loads are imposed, a certain percentage of that load is transferred down the axis of the leg (and trigonometry/vectors can be used to determine the numbers): the remainder of the load that is not resolved in that manner continues of course to try and push the leg sideways. The leg, as it is on a slope, cannot move sideways easily – usually the friction of the foot at the floor will give. If the foot was fixed however, the leg would try to rotate, with the floor as the pivot point. This rotation would mean that the upper end of the leg would need to move upwards of course. So, a heavy beam on top of that sawhorse, the same beam that might be inducing those lateral loads, would also be serving, on the basis of its weight to resist the leg wanting to rise upward as it rotated. Thus the splayed form is relatively self-stabilizing.

    A corollary in timber construction would be, say, the long sort of braces employed in German traditional wall frame systems. With the addition of a metal tie rod to resist any upward movement (a tension), or simply the weight of the building carried by the wall, any lateral load imposed on the side of the wall, will have its load transferred through the long brace directly into the ground.


  9. It's big, just when I concluded that you used only a limited tool-set, I am confronted with an impressive toolchest.

    I would add some kind of lenghtwise slot through the centre of the table to also allow clamping there and not only on the sides of the table. Although with a thin top there is an alternative, as it is possible to use and insert adapted clamps through small holes. Festool for example inserts clamps through 20mm (3/4″) holes on their MFT tables series.

  10. Damien,

    now some woodworker in Indonesia, I'm sure, would be envious of my 3-chisel set. Whaddya talking about – “limited tool set”?!

    Well, while I suffer from a lack of stationary power tools, my set of hand tools is quite extensive and the fact is I have only shown a small slice of them doing the various builds on this blog. Also (and I know you were probably ribbing me, but all the same), I plan to store hardware and other stuff like that in the bench, not just tools. I'm also thinking about how I might store long clamps with the bench, for example.

    Your idea of the lengthwise slot is a good one, and funny you mention it as I was remarking to my wife this morning that one advantage of a sawhorse is the relative ease of clamping certain things directly to the beam, and how some way of making a bench in some similar fashion, like having two beams side by side, might be an idea. Hmmm, wheels are turning here.

    I'm not sure if the Festool clamps are the ticket with my 'thin' 1″ top, however there are options, and I'll given that some thought as the bench design mutates.


  11. Chris

    As I have 2 Makita clamps similar to the Festool, I can look if they can be inserted in a 20mm by 25 mm (1″) thickness hole. But I think that the Festool table top is also 1″ thick, 1 as it needs sufficient stiffness for clamps, 2 a replacement top on Amazon has that thickness and 3 as it is a sacrificial top to be cut 1/4″ or more deep. Anyway I also have to look for solutions for my own clamping problems, and cutting large slots in my tabletop, generates structural problems.

    As for the extension of your tool collection, I am mainly curious about what is used in your projects. I thought it could be limited, but it is hard not to accumulate over the years. In my case I was only able to restrict the size of my library, as I set the total shelf length 15 years ago.

  12. Hi Chris, your comment “I intend to design the piece to satisfy my anticipated work needs, however I have been able to find ways to use a sawhorse for most everything so far, so I can't help but think that the bench will add a lot of capability in regards to fixing the work in place.” Reminded me of the last message Chris Swartz leaves you with in his book.
    “This is…exactly how I feel when I'm working at my workbench. Workholding is easy, leaving me to worry if the surfaces are true or if my joints fit. 'I almost never have to stop and ask: How will I hold this part to work on it? This freedom makes me less of a jigger and rigger and more of a joiner. And I like that.” I really commend this book to you.
    Were it me, I would leave a space under the top for holdown/dog clearance, a space under the front for your feet and I would make the front-side structural members plumb and flush with the edge of the benchtop. This will greatly simplify the features you will need to clamp boards for edge work. There are very few times when you will generate a wracking force towards you, so what you lose in stability will be greatly outweighed by improved holding facility.

  13. I checked the (Festool like) Makita clamps. Inserting a clamp in a 19mm (3/4″) hole through a 22mm board was problematic, where a 22mm hole worked fine.

    A workbench alternative I liked was Joshua Finn (fww202) approach with two big sawhorses (16″ 48″ and 32″heigh) completed with two torsion box beams (9″ 4″ 8') with melamine on one side and Homasote(?) on the other. One side for gluing the other to have a sticky surface to make clamping nearly unnecessary and avoid damage. I am probably going to reuse some of his ideas (maybe cork?).

  14. The top I am gravitating toward right now is…
    A slab 6'+ as wide as the planer can handle. No sense in creating maintainence work. Stuff it in the machine. So for me today a humble 15″ width. But I see a tool tray in the back side with the farthest side as High as the slab is as to create a work support. Creating a work surface of 24″+ wide. The closer tray rail that would touch the slab top i would make a 1/2 lower to be able to use it as a shooting rail right off the side of the slab.
    Though I am happy with the trestle horses (never have posed any problem in racking) with a simple plank below to store what have you.
    What I find is that though I may have a big bench surface, I used a fire rated door slab with oak veneer for several years and never broke through the veneer even with aggressive scaping and cleaning. I did alot of work on it too. So, I end up using mostly this one spot at the end where I do most all cutting and chiseling of small to medium size work. So I started thinking why do I need to go bigger?
    Your idea about two rails is side by each is interesting. Maybe two planing beams with a narrow tray between. If they were 3x12x72+ then face jointing and resurfacing with a machine would be a snap.
    Now, is this in addition to the kitchen table or do you intend on relacing it with the bench? Or expanding the floor plan of your kitchen? 🙂
    C on da big

Anything to add?

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