Wednesday, July 20, 2011

Coffee Anyone? (17)

Post 17 in a series describing the design and construction of a coffee table, with previous installments to be found in the blog archive to the right of the page.

In the last post, I completed the work to join the table top frame members together. Next up, I had a few loose ends, joinery-wise, to attend to.  The draw bars which connect the legs to the table shelf frame needed to be fitted up. I sized them so that they were a mild interference fit in the leg mortises. With the bar driven in with the aid of a mallet, I used the hollow chisel mortiser to prepare the fixing pin mortises:

A short while later, all four legs were fitted with their draw bars and mortised for the pins:

The next item on the 'tick list' was to trim the draw bars to final length. I decided to modify a jig i made previously so as to allow the legs to be fixed identically and run through the table saw:

Here are the four legs after the table saw work is complete and the draw bars shortened:

I then fitted the leg and its draw bar up to the shelf frame pieces:

A clamp ensured that the joint was fully closed up:

Then I used a hollow chisel mortise bit to mark the draw bar:

 I separated the pieces, then set up a small kama-kebiki to mark the mortise lines on the draw bar, bringing the lines in 1/32" or so from the mark left by the hollow chisel bit in the previous step. This allows for a draw-bore effect:

Some mortising took place after that, using a small drill bit to punch out the hole, and then a chisel to clean up the mortise:

 The pin mortises have slightly less relish on them than I would prefer, however I believe they will be fine considering the hardness of the material and loading conditions on the joints being not especially high. To obtain the ideal amount of relish, I would have had to widen the shelf frame pieces by at least 1/16" (1mm). Next time.

Mortising work complete, I then drove the fixing pins, komi-sen, into position:

Pin after it has been driven in:

Next, I trimmed the pins flush with the surrounding surface and cleaned off using a pass with a paring chisel:

That pretty much wrapped up the joinery work, once all four legs were similarly complete.

I then commenced the finishing work, which involved a fair amount of chisel work on the leg 'stirrups':

Then it was a matter of more chiseling, then some scraping, filing, and sanding, over and over again, to complete the legs. I didn't take any photos.

It was at last time to apply oil, round 1:

Here's my drying rack:

My apologies for the blurry pictures - the studio photos I'll get done next week will hopefully make up for it.

And round 3 of oiling for the Wenge shelf panel:

I experimented with using the Whey based finish, but I decided I preferred the darkening imparted by the Tung oil to the bubinga, so I'm sticking with Waterlox for this project. Otherwise, the whey-based finishes were fine, no VOC's and easy to apply, rapid drying, etc., and the next project I do involving a lighter wood, like oak, say, I will likely go with them.

All for today, thanks for coming by.  --> on to post 18


  1. Chris, I think this the first build that I understood completely, including the rationale behind the joinery. Are the shachi-sen always employed in pairs? I was considering how I might use them in a future project.
    Nice work as usual. This table has a particularly pleasing shape to my eye.

    Harlan Barnhart

  2. Chris, although I don't comment on every post, I follow religiously. I'm constantly amazed as the precision of your joints and the complexity of them. Do you ever come to the NW to teach? If I had enough notice, I'd be able to get a week off to attend a workshop with you. I'd love to get on hands tutoring from you.

  3. Harlan,

    glad it's making sense to you. Shachi sen can be employed in any number really, though paired is common, or one on each end. There are western cogged beams which use an equivalent of shachi sen to lock the beam halves together, the pins spaced every (say) foot or so.


    Nice to hear from you. I haven't been to the Northwest in a while, but always like to travel out that way when I can. How about this: if you want to organize a workshop for at least 6 people, somewhere out on the west coast, I'll come out there and teach. Please Email me if you want to follow up on this idea.


  4. The best I could do is either try to talk Gary Rogowski (Northwest Woodworking Studio in Portland) or the Port Townsend School of Woodworking to host you for a class. I could get away to either of those. I don't know of a venue better than either of these anywhere close to me.

  5. You don't often see the tenon grain direction as you have done there at the tops of the legs. It seems that having them run in the stronger direction and bridling the draw bar would be more conventional/stronger. Not enough space would call for a design change perhaps.

  6. Hi Vic,

    have you thought about hosting it at your shop? Or a friend's shop? Thanks for thinking about the options.


    I must say that your comments and criticisms give me frequent opportunity to look again at decisions I have made and to work on my equanimity, and in choosing a helpful way to respond. I thank you for that.

    In the case of the legs, the first factor governing the grain orientation, besides the wood available at the supplier, is the appearance of the leg after cutting. The bulk of the leg is rift grain so that cutting it out into a circular shape results in a quiet grain appearance along the front and sides of the leg. At the upper end of the leg, the grain turns toward a flatter orientation, which results in the grain running across the short dimension of the tenons.

    This brings me to the second consideration- joinery is ideally configured so that the grain on tenons runs across the short dimension, not the long one. This minimizes movement in service, and considering that the tenon is sandwiched between the table top frame members, I wouldn't want the grain orientation the opposite, as you seem to suggest, as this could present the opportunity for tenon swelling to force the miter between the frame members apart.

    As for 'bridling the draw bar' - that is exactly what I have done. Are you sure you are understanding this connection correctly? I realize that some of my photos are blurry and that might convey an imprecise picture.

    With the tenons turned 90˚ from their current configuration, as you appear to be suggesting, then there would be no opportunity to use a draw bar as the tenons would pass right through the bar.

    As far as 'strength', goes, I am puzzled by this in regards to direction. I am guessing perhaps you are thinking of resistance to shear parallel to grain? If so, tests have shown that there is no statistical difference in strength in shear resistance in relation to the angle of the growth rings (see: )

    in any case, the only load I would worry about here are simply the dead weight of the legs and shelf, if the table were picked up by the upper frame- a trivial load to worry about.

    And in timber framing, there is no Japanese protocol I am aware of that specifies growth ring orientation in regards to tenons. Perhaps you have some info you'd like to share?

    If you feel you have a better joint design for a situation identical to the one I have, a joint that does not employ glue or metal fasteners, I would be intrigued to know what it is.



  7. Sorry Chris, i didn't mean to tax your equanimity, you obviously have enough to think about. You defend your methods quite well. My take was not questioning having the grain running tangentially across the tenon as I believe that you have done there, and is indeed generally the preferred method, especially with the mortise cheeks running the same way. It simply occurred that having the tenon length running parallel to the curve of the leg, rather than perpendicular to it, would have been the more standard choice, but I could be wrong and not seeing something. Still bridling the draw bar would end up with you having four stub tenons, a possible loss of strength unless you half lapped the bridle or something.

    Perhaps a bit over complicated, and you are working with limited space there and complexed by the draw bar, which is what I was referring to when mentioning having to possibly alter the design. In any event, the tenon shoulders carry the weight, so as you state, the strength concern is rather limited.

  8. Hi Dennis,

    thanks for your reply. If I ran the tenon lengths "parallel to the curve of the leg", they would stick out of the top end of the leg more or less in horizontal alignment - then I would have short grain problems with the tenons.

    The grain in the legs is running perpendicular to the ground, and therefore the tenons are aligned to the grain's run.

    If you could send me a sketch of what you mean, perhaps that would clarify matters. I'm either misunderstanding you, or you are not quite seeing the connection clearly. In either case it would be great to obtain some clarity.

    Your input is always appreciated.


  9. Replies
    1. Paul,

      thanks - appreciate the comment!



All comments are moderated:

Anonymous comments will be deleted

Spam will be deleted

Comments containing links unrelated to blog content will be deleted