The last couple of days have comprised some final fitting and finish planing of the parts. I had one patch to do in the side of the top beam:
The Canarywood is fairly cooperative when it comes to dragging a blade across its surface, save for the odd minor patch of rowed grain here and there. It's a bit like Honduran Mahogany in that respect, only harder and stinkier. To tell the truth, I didn't fuss overly with the planing, knowing how banged up this sawhorse will be in no time, and only resharpened 4 or 5 times during the process:
I also chamfered all the arrises, though it will leave tiny gaps here and there when certain pieces are brought together, but I'm not bothered about it:
The glue up, though the epoxy gave me 30 minutes of working time, was a bit of a nail-biting stress session. Actually, I didn't even have time to bite my nails and was fortunate to have my wife around to help hold some part together during the first few stages of assembly. I might have been better to use urea formadehyde glue for its long open time, but I wasn't sure how well it would bond to the Canarywood and didn't feel like buying a $10 box of it for sake of experimentation. Here's the horse all glued, clamped and wedged up:
I decided only to wedge the legs that are à devers, as the tenons on those two were shaped in a conducive manner for wedging. The faces aplomb legs with their irregular pentagonal tenons seemed too much of a hassle to wedge as the wedge would be working against the grain in a poor manner and I would have to make a parellelogram-shaped wedge for one end. Nah...not this time:
In the next picture, you can see how I dealt with the barbes that taper out to nothing - I trimmed them back about 5/16" or so:
The epoxy was dry a couple of hours later, and I could start trimming the excess off:
Here's the other end:
So, after the dust had cleared and the pulse rate settled down a bit, here we have a nearly-complete project:
Another view - it isn't the easiest thing to photograph, as every photographic angle has some things in view and some things hidden:
The Canarywood sure is a nice looking material in some places - here's my favorite spot:
Okay then, all that remains is to fit the sacrificial cap, and that will be on the slate for tomorrow, so I hope to see you then. I found a nice piece of radial grain fir for the cap. Thanks for dropping by today.
--> Go to post 31 (final!!)
Quite stunning, Chris. I have continued to follow all of the posts, often unsure of exactly what I was looking at due to the incredible completxity of the joinery...and even now, when I see pictures of the completed assembly, I have a difficult time visualizing it...as strange as that sounds.
ReplyDeleteDo you think that you would have been able to build it without first having modeled it in 3D using CAD (ie using traditional hand drafting techniques)?
At any rate, what a wonderful "craftsman's sample" that you carry with you to show prospective clients!
Hi Jeff,
ReplyDelete"Do you think that you would have been able to build it without first having modeled it in 3D using CAD (ie using traditional hand drafting techniques)?"
Good question. I started out about a year back and spent many, many hours drawing the piece in 2D using a software called MacDraft. Without the benefit of an actual example to look at, and only the book's drawing with it's numerous mistakes to guide me, I was swimming in the dark in several respects and having difficulty visualizing some of the connections. I reached a point of about 85% completion of the drawing using 2D alone. Of course no one was there to teach me either, though I had tried asking two different campagnon for help (one was too busy to help and suggested I bring him over to teach a course in he material and the other lacked the book and couldn't piece it together from the information I sent him).
Based on having 2D only to proceed with, i could have cut the sawhorse out and would have had some pieces fitting and some pieces not, so it would have taken me a few go rounds of cutting out the horse before I did it correctly.
Then I tried SketchUp about 8 months back, which was new to me, doing what I could using 2D on the floor of the drawing and then developing from there into 3D. That was a godsend as it allowed me to both confirm that much of what I was doing was in fact correct - you can tell in 3D that when a series of points an lines are true, a plane is formed automatically, and it allowed me to see parts of the sawhorse and how they connected much more clearly. Not everything of course, for in the end i was caught out by that impossible tenon issue, and thus have some mortise scars on the top of the beam at the end.
Still, 3D notwithstanding, I had to repeat and repeat the drawing many times before I had all the parts worked out, and throughout that process I stuck to the method of drawing in 2D on the floor and then projecting up into 3D, so I would not become overly reliant on the 3D. One can see that it would be esy to get a little lazy with the 3D, however 2D drawing is an important carpentry skill and I don't want to shortchange myself in that regard.
When it came time to take measurements for cut out, I took them off the 2D part, and did a simple confirmation on the 3D to be sure the measurement I was taking was the correct one.
~Chris
A few question and comments Chris:
ReplyDelete-First off I've been looking forward to seeing this go together ever since you mentioned long ago that would be one of your projects. It's pretty sweet.
-Second, it seems like it would be both more intuitive and accurate to print out your 2D drawing and pick the joinery directly off of the drawing. Why did you go with measuring all of the joinery intersections?
-Third, it made sense to me from a practical standpoint, but it struck me as a bold move to kerf your joinery intersections. I know from experience that adjusting anything compound is far from intuitive, where you fix one thing, and it screws 14 other things. Were worried about that, or is your Yataiki saw only 6 microns thick?
-Lastly, on the assembly issue, it's always struck me that the crazy French sawhorses are exercises to better understand roof framing (and maybe to impress fellow compagnons, perhaps even girls). They would draw the theoretical through tenon where it pops out, even if there is no tenon. If it's a drawing/roof framing exercise, then the fact that it is impossible to assemble as drawn makes sense. That still doesn't explain the errors in the drawing.
Thanks,
Brad
Hi Brad,
ReplyDeletePrinting out the drawing would have been one idea, however I don't own a printer and besides the drawing is way too large to do on anything other than a commercial plotter - in fact, it's really too large for that, unless each view in the drawing were detached and done separately, a potential source of error. I did consider drawing it full scale on a door skin (which still would have been too small - two door skins would have been needed, or possibly three. However I am working in my kitchen at the moment, a cramped space with a very wonky floor and I just don't have the room for that, nor the place to store the door skins afterward. The method I used was accurate enough, of course anytime you are taking numbers off from one place to another there is potential for layout error.
The kerfing seemed to work well, though I proceeded cautiously. I would say the results are decent, but, of course, could always be better. Some of the interior joint angles are so awkward to cut that kerfing in with a saw actually gave the best results I believe as I couldn't get even my skinniest chisels into the very acute corners.
"...it's always struck me that the crazy French sawhorses are exercises to better understand roof framing...".
Well, I have pointed that out on the thread many times, just to fend off the brigade of people who think I'm crazy for putting so much effort into a 'lowly' sawhorse. Your explanation for the tenons might make sense in certain contexts - for instance in the case of a Mansard lower pitch hip corner and the use of a single brace, which could be slid in with the shape of tenon that the drawing indicates, however the fact remains that is an impossible assembly when the brace is lapped to another one. And is it also true that the perspective view illustrates through tenons for parts where it simply is impossible no matter how you slice it (excuse the pun). That makes no sense other than to be deliberate thing- why go to the trouble of illustrating it, and why make no mention in the text about the 'theoretical' nature of the tenons in certain places? Besides, the leg tenons which go through the top beam are mis-illustrated in two spots anyhow.
I wonder if the illustration was done by several people who didn't communicate much, or who didn't understand the drawing in depth? Who knows - I remain a little irritated at the book though, and am less excited about getting into more parts of it, though the disillusion may fade in time.
By comparison, I have more than a dozen Japanese layout texts which I have studied in exhaustive detail and I haven't found a single mistake yet, so I have great faith in them as a source of information and guidance.
~Chris