French Connection 6

I thought it was time for an update of my work on the Mazerolle tréteau. Those readers who are new to this blog and haven't come across my previous posts on this topic, please refer to the label index to the right of the page and you should be able to track down the previous installments.

When I first posted up about this sawhorse on the blog, I was at a point where I was able to solve some of the drawing, but there were a few areas of it which eluded me, largely because it is fairly tough to visualize the connections in a piece like this, where each of the four legs is in a different rotational arrangement in relation to the top beam and all the braces. I am also coming from a background in Japanese carpentry drawing , so the French methods are a little different and more than a little head scratching goes on as I work to get my head wrapped around it. It gets easier with time - that, or my derangement grows, I'm not so sure!

Later on, with the help of SketchUp, I was able to develop the 2D into a 3D model, which allowed me to confirm the layout, and to finally get a good look at all sides of the piece. The book shows a perspective view from one corner only, in this is in fact mis-drawn in one area, so it was a struggle to imagine how the connections came together on the backside of the piece.

Once I had the drawing complete, I scaled it down to produce a sawhorse at the height I desired, which, in this case, is 24". That was how I left off the previous post in this thread, with the drawing complete and scaled.

The scaling-down had brought the piece to the desired height alright, however it left the sizes of the pieces of wood with some odd sizes. The interior X-braces that sandwich between the long main exterior brace sets were less that 3/4" thick, which was clearly a bit scant.

Of course, in studying this piece I initially followed the drawing in the book as faithfully as I could, at times with the aid of a magnifying glass and scale rule, and drew the initial versions in metric. I wanted to produce the 24" sawhorse with inch-scale components however. Then, when I set about re-drawing the horse with components at the desired sizes - the legs for example will be 2" square sections - I found I hadn't understood the drawing as fully as I had thought, as I ran into a few glitches. Trying to get those interior braces a bit thicker was not so simple as I had imagined.

So, another flurry of study ensued, and another 20 hours of drawing later, I think I have a somewhat better understanding - certainly enough to build the sawhorse at this point. Now I can specify the sizes of the components and vary the slopes of the lags and braces, and have some idea what the ramifications of these changes to slope will be. One thing about a horse like this is the utter inter-relation of all the parts, so a change to one dimension ripples down through the rest of the geometrical relationships, sometimes causing problems. A complete understanding of this piece will come after I've built a few of these, so it's years away at this point.

While I still have another 6~10 hours of drawing to resolve some of the remaining mortise and tenon connections - connections that vary on the sawhorse in detail depending upon location - I thought I would share with the reader some recently-completed drawings. I've illustrated the various pieces with different woods - Mahogany for the top and legs, Ash for the long braces and the short braces, and Purpleheart for the interior braces -I did this as I thought it makes it easier for the reader to see what it going on with so many different pieces involved. These are not necessarily the species I will use - I'll see what I can find later this week when I go trawling for some material. I want wood that will be medium weight, fine grained, decent workability, and good rot resistance - and not too expensive. I was hoping to use Black Locust, but I can't obtain any dry material, so that would appear to be out. That's a bummer, but I'm sure I'll find something. Uh, well, maybe not sure, but at least hopeful.

Here's some snapshots then.

Top View:


The illustration in the book shows the mortises in the top beam drawn incorrectly in three cases. That kind of thing doesn't help when you're trying to solve such a riddle. I'm imagining this drawing was either made by an apprentice who was copying from another drawing, or the mistakes are deliberate and meant to confuse. When this sawhorse is complete, the top view will be a little different as I will be fitting a sacrificial cap as I did on my other sawhorse (see the posts "Irregular situation", parts I through VI)), which will cover all the exposed tenons.

This perspective view is fairly similar to the one drawn in the book:


A view of one of the short ends of the horse, here the particulars are that the the left-side leg is rotated 45˚ to plan, while the right-side leg is aligned to the long axis of the plan:


One of the long braces extracted (part of the assembly termed the croix de Saint André dans les liens Mansards):


I'm going to do half-laps with mitered abutments as per usual, and in the above drawing the tenon at the lower end is not yet defined.

A view of the other short end of the horse, where the situation is different again - the left-side leg is at 45˚ to plan, while the right-side leg is aligned to the short axis of the plan, and I have removed one of the short side braces just to give a look:


I'll finish off with a view of the spaghetti junction - and not all the lines are yet on this drawing, As I haven't developed the rotational views of some of the braces yet:


Click on the picture and it will all become clear :^)

So, I hope to commence work on this sawhorse within the next week or so, and the next post in this series will be the commencement of the build thread for the piece. I anticipate that the upcoming tréteau build thread will not be so long as the previous build thread on the lantern. That's the plan anyhow, and hopefully there won't be too many surprises in that regard.

Thanks for visiting today. On to the build! <-- link