Thursday, February 18, 2010

Tréteau XXV

After my previous post, I received the following comment from Woodjoint:

"I just keep wondering how it will all go together in the end but I'm sure you've thought of that as well. I'm always putting the piece together as I'm building it so as not to have any surprises in the end."

Not the first time Woodjoint has mentioned that, and that's okay. As I've mentioned several times during this build, one of the unknowns for me has been in fact how the piece will assemble. This build has been an exploration in the French devers de pas layout method, and as I pointed out many posts ago, there are quite a few errors in the original drawing by Mazerolle which I've already discovered - 10 or 12 of them in fact - and finding those goof-ups had caused me to wonder about the drawing in several respects. Some of the 'issues' could be put down to errors by the engraver, however other ones are not so readily explained. Anyhow, I have made the necessary corrections as I've gone along, and blundered quite nicely into several blind alleys, including the one mentioned in the previous post where I made the error of referring to an old drawing which had something very wrong with it, and thus laid out and cut 8 mortises all wrong.

These, I guess, are the prices to be paid for such explorations into the relative unknown, and yesterday I came to another ka-ching to be applied to the grand total, when I found that the long side braces, when assembled to one another, really didn't want to fit to their mortises. Everything had been laid out and cut 'correctly' - at least as far as [my understanding of] the original drawing had indicated - the problem was that the parts simply will not go together in that manner. This was an area I had been concerned about from the get-go, however I had placed a certain amount of faith in the Mazerolle drawing, and figured that I would see when I got there, and that perhaps the parts would go together in some partial rotation, or some other manner that wasn't entirely obvious. But no, that does not seem to be the case now that I have tried it with the braces assembled to one another. It's a little deflating, to say the least, to find this out at this stage, after so many hours. And there was no ready way to find out otherwise- while each individual brace can be fitted, when the braces are lapped to one another they cannot be fit as a unit into the top beam.

At first, when things started looking like a no-go, like many people, my first response was denial. That can't be!

Well, it was.

My next step was to look again at the book to see if there had been something I had missed. Well, okay then, let's take a look all together shall we? Tell me if I am imagining things here. The drawing in the book that shows the development of the long side braces and their tenons is this one:

The view is of the inside faces of the brace pair, and the line from 'R' to 'T' is the top edge of the sawhorse main beam. As pointed out in a previous post, the developed view of the brace above is actually a mirror image and impossible, however, one can see in any case the trace from the tenon on the developed view back to the view of the inside of the brace pair. Notice that the tenon, near 'T' (and near 'R') does not angle from the main beam at 90˚, but rather is shown at an angle relative to the brace itself, tapering, it would appear, about 50% from base to tip. Notice also that the brace is clearly shown to have a half-lapped connection.

The shape of the tenon is shown also in the main plan view of the sawhorse, and in the next picture I'll zoom right in on it, looking here at the top of the beam:

The tenons near 'M' and 'P' are for the legs, and the next pair over from them to the right in the picture are the tenons for the long side braces. If you look closely you can see the dotted lines which trace the shape for the tenon from the underside of the beam up to the top. Clearly it is a through-tenon we are looking at, and clearly it seems to diminish from root to tip so that the exposed end of the tenon is about half the size of the base. The trace of the entire brace is shown on the top of the beam and you can see that the exposed tenon is only half as long as the outline of the trace.

Well, that is what the book shows, and that is what I followed, however, as I mentioned, with the parts now more or less cut, it is looking like these parts simply will not assemble to the top beam. Fortunately, in my second go-round with making these braces, I haven't yet cut the mortises in them for the little interior x-brace sets. I say fortunately, as I'm now wondering if these will assemble as well. Again, here's the original drawing:

Trace the line from the developed view of the brace tenon at the bottom of the picture and you will see that it is indicated to taper about 50% from base to tip, and again the braces are supposed to be half-lapped together. Hmmm. Now I'm paranoid.

So, things have certainly gotten interesting in this build. What to do? Here are some of the options as I see them:

1) burn - crush -destroy entire sawhorse

Okay, well that's not going to happen, but the thought did flicker across my mind!

2) Open up one or both of the mortises in the top beam to allow the tenons to slip by, and then plug/patch the resulting hole(s) afterwards

3) taper the tenons much more radically so that they can be slipped into the mortises.

4) Some combination of (2) and (3) above, open the mortises a bit, taper the tenons a bit...

5) Cut most of the tenons off, patch the mortises on the top of the beam, and use timber screws

6) change the joint that connects the two braces together from a half lap to a joint involving three parts.

I'm not wild about option 5 at all, and 2, 3, and 4 are hardly what you might call appealing. If I were to only taper the tenons to the point where they would slip in easily, then I would be cutting them to the point that they would be about 90˚ to the top beam, and it looks like such a re-tapering would make them so that they wouldn't even reach the top of the beam, or come to a point at the top of the beam.

At this point I am leaning towards a reconfiguring of the lap joint, option 6. Here's where the error of having mis-cut those brace mortises previously may have a bright side, as I can make use of the ruined pieces still to construct new joints, and I can think of several good options for a different joint. Using a different joint should allow me to keep the main beam mortises and brace tenons as they are currently configured.

As for the interior x-brace connections, well, I'll worry about that later. One thing at a time.

Well, this situation sucks pretty much any way you slice it. Funny how when the end is in sight you suddenly hit a pothole. I can probably find some solution to the problem, and it will come out okay I'm sure, but again this leaves me wondering about the drawing in Mazerolle's book - given how it clearly shows the tenons to be of a certain form, and given that the sawhorse will not assemble when the parts are cut in that manner, what gives? What's going on here? Is Mazerolle laughing in his grave? Was this drawing designed to mess with my mind, render this carpenter a drooling idiot after having attempted the piece? I don't know the answers to those questions, but I'm a bit frustrated and more than a little annoyed at this point. I keep hoping some some proverbial penny will drop, and I'll suddenly see some little detail I had previously missed or incorrectly interpreted, and have an 'aha!' that will make sense of the puzzle, causing everything to fit together without issue. But, after hundreds of hours staring at this drawing, over and over again, not to mention drawing it for myself umpteen times, I'm unable to locate any little piece of info that will rectify the problems I have found.

I'll examine this problem some more today nonetheless, to check again if there is something I may have missed, and reconfirm again that the pieces really won't assemble (I'm still in denial) before I decide upon the course of action to repair the problem. I might play around with the drawing some more, god help me.

Well, this will mean a slight delay in the build, but notwithstanding the aggravation at this stage, it's all good learning in any case. I couldn't have learned all I have about French layout technique (which may not be much at this point) without at least having tried to make this piece. Nothing ventured, nothing gained.

Thanks for your visit today. --> Go to post XXVI


  1. Sorry to hear of the setback - I've been eagerly following the build, anticipating the final assembly.

    I'm not entirely surprised that errors can come out of working from a 2D drawing to build a complex 3D structure, but I don't understand how these issues weren't evident when you modelled it in Sketchup... surely a fundamental problem with the drawing would show up in the model.

  2. Hi Daniel,

    ah, well that's an interesting point. It is possible to draw in 3D, in SketchUp or any other software like it, all sorts of objects that cannot actually be assembled. It would only become apparent if the drawing were animated and one could move component parts in and out of one another while observing what happens at the joint interfaces. Now, there are actually some animation functions with SketchUp, especially if one buys the Pro version, and I could have dragged components in and out of one another to see how things work, but, then, that would have taken a bunch of time and it didn't occur to me to do so at the time. Hindsight is 20/20 of course....

    All that said, there is one important point: a perspective view of the sawhorse is also shown on the top of the Mazerolle drawing, and in that view it clearly shows the interior x-braces having through-tenons into the long side braces. I've already realized this facet of the construction is impossible. Therefore, the drawing in Mazerolle's book, whether we are looking at one development view or another, or in the perspective view, shows an impossible construction in regards to the long side braces and interior x-braces and their connection to each other and the main beam. I've stared for a good while this morning at the drawing again, and the conclusion seems inescapable. I have some ideas why this is shown like it is, and I'll share them in the next post.



All comments are moderated:

Anonymous comments will be deleted

Spam will be deleted

Comments containing links unrelated to blog content will be deleted