Previous Posts in this French Connection series: 1, 2, 3, 4, 5, 6, 7, 8 (<-- each number is a link). A friend of mine from California, a certain Mr. Laine, sent me some pictures of various French carpentry models he had seen on a trip a few years back to Paris. While I ended that past build on the Mazerolle Tréteau grumbling about various things and definitely feeling less than enthused in some respects, these new pictures rekindled my enthusiasm, a couple in particular:
That piece could have a role in an Alice in Wonderland film. I get vertigo just looking at it. Take a look at that support table for starters!
Here's the other one:
I wonder if there was some possible form of complex construction which they left out of that model? I doubt it. One of the parts I like the best is way up at the top, where we see some Saint André Crosses in double curvature helping support a balcony:
So, a bit re-enthused, I decided to plunge back into some more study, which means drawing drawing and more drawing. Way back, it might have been the first post in this series, I showed a roof from the Mazerolle book which appeared to be a great application for the bracing methods employed in the sawhorse study:
So, I've been working on drawing that structure. In the book, the roof actually sits atop a parallelogram-shaped plan, which may not be apparent from the above perspective sketch, however I opted to first develop it on a regular rectangular plan. the odds of constructing something on a parallelogram-shaped plan are rather slim, I would think, I can learn the drawing techniques just fine on the regular plan.
I'm far from finished at this point, but thought I would share progress:
I'm only just starting on developing the braces in the lower section of the roof:
There are several curious things about this Mansard roof compared to most Mansards you will come across. The end wall, instead of being two pitches like the long side, is actually a single pitch:
However, it's made more complex, by far, due to the fact that the end wall roof pitch is not the same as either of the pitches on the long sides. The steep pitch on the long side is about 30/10, the upper roof on the long side is 5/10, and the end wall pitch I have set at about 12/10. That makes for some design difficulties where there areas collide, let me tell you. All the hips are compound irregular, which, in rectilinear roof work, is about as hard as it gets. I like a challenge though. I like investigating stuff. My wife tells me I am a number 5 on the Enneagram model, "The investigator".
At first, the roof appears to follow the same protocol as the tréteau in that the lower hips are each in a different rotation. This picture captures a sense of that:
The hip on the right side of the picture is rotated so that one of it's faces meets the long side of the plan flush. Note on the furthest jack rafter to the right the long barbe that results. Fortunately, there aren't too many of them to deal with. The hip is not square, but rather is a somewhat rectangular section.
The hip on the left at first appeared to be a case of being rotated to the short side of the plan, but later I discovered that it is not quite all that it appears. It is in fact akin to a regular sort of backed hip rafter that has been sliced in half along it's length. It is therefore a rectangular section with a bevel (the backing) taken on the top surface. Why do they do this? Perhaps, as a local suggested, it is something in the wine of the Alsace region? kidding aside, while I find it an interesting variation on a Mansard, I'm not entirely sure I like the shape. We'll see how it looks when there are panels on the roof to clean up the look.
Well, the reasons for the odd hip rafters on this roof are a little complicated to explain in just a few words, but one point involves the matter of the inside surfaces of all the rafters and the planes of intersection that result with irregular hip corners. Here's a view of this half-sliced conventional hip, from the inside of the structure:
You'll notice how both jack rafters that flank it meet the underside arris cleanly. That's the main reason, I do believe, why they do that with the hip. If the hip were a rectilinear section simply rotated to meet the short side plan, the jack rafters on one side would stick out below the edge of the hip. So this model is not strictly an application of the sawhorse problem, it adds more stuff to vex the student. bien sur, c'est Mazerolle!
The roof structure is based on this basic truss:
I'm a fan of trusses, as triangulation, and the transmission of loads along the grain and through end grain, makes for high strength with minimal amounts of material. These trusses are of the king piece type. Here's a view of the inside of the structure showing some of the truss configuration, as I currently have it sorted:
The trusses employ a doubled lower chord. The cross bracing will take a while to figure out, as it differs from the sawhorse in that it is asymmetrical. Also, there are more braces to go in, which are like those in the main truss, but travel from the hip corner horizontal braces to the hips themselves. This will be something to scratch my head over as well, I'm sure.
If I ever get to the point of owning some land again and can build myself a shop, I have long thought that it would have a Mansard roof. It is a very strong and practical roof form. Whether it will be this version of Mansard roof is hard to say. Mansards are better suited to stone walls in some respects, as that allows for little to no eaves. With eaves, a slack third pitch is required. I think I might add such to this drawing once I'm through with the brace sorting, probably a curvilinear up-swept eave, and am also having an urge to put an eyebrow dormer on one side of the upper roof with a single-curvature Saint André's Cross fitted into the rafter structure. That would be too cool by far!!