Monday, February 22, 2010

Tréteau XXVIII

Post 28 in the build, approaching the end. Previous installments archived at the right of the page.

I'm still more than a bit puzzled as to the outcome at this point. The only thing I can conclude, that makes even the remotest sense, is that the drawing in the book is deliberately misleading - I presume the reason for that being the fear that the text, in the time in which it was originally published, would fall into the hands of a competitive carpentry school. And if that is what it is designed to do, then it fails in the end, as I have figured the method out, and now will be a lot savvier with working on any further examples from the Mazerolle book. I'm now expecting to be misled. And as for whether I will work on further examples from that book, it won't be for a while. I need to lick my wounds.

Now that I understand the method of layout presented in the text fairly well, I can stand back and ask if the sorts of things it facilitates are in fact the sort of things I need/want to know. I'm not so convinced about that point any longer. The adding of x-bracing, the Saint André's Cross to a splay-posted structure I am feeling is largely unnecessary. While it may indeed add a certain amount of rigidity, the splay-posted form is inherently quite stable as it is, and the added complexity of trying to get a multitude of angled pieces all to come together at the same time is a tremendous added difficulty and time burden.

In short, from a paying point of view, if I might be so crass, since the cross adds mostly cost, little apparent benefit, and the assembly itself would be concealed in the roof structure, why would any client want to opt for it? If it adds extra stiffness that is largely un-needed, at the expense of a lot more hassle, why would I want to do it as a carpenter, other than for amusement? Further, the x-braces, if they are to be incorporated into some single-layer hipped roof, leave one with much added work in terms of fitting jack and center-common rafters in and around them - again, a lot of added cost and hassle for negligible benefit. Finally, from a purist joiner's viewpoint, angled braces are always problematic anyhow because they function only in compression and any pegging is, at best, only of aid in assembly, because the angling of the brace results in a tenon with precious little relish beyond the peg. A moderate tension load will blow the relish out. It seems the French more often than not in this form of construction forgo the tenon anyhow and simply spike the braces in place, which is a dubious connection from my point of view. Even when they are using a tenon, it is invariably a bare tenon connection only, which is considerably weaker than a shouldered or housed tenon. Some drawbacks to be sure. Well, perhaps I'm just a bitter old man, mumbling and grumbling by this point and am missing out on some wonderful and important quality that the Saint André's Cross imparts to the roof structure, a quality that outweighs its many drawbacks, and if any reader would like to set me straight on this matter, I'm very open to hearing about it.

So those are some points I wanted to make now that I have reached this point in the process and the bubble has burst on me a bit as far as the overall methodology in the Mazerolle book. I am still grooving on the layout method though, and in fact in the workbench I am currently sketching, which will not have the angled braces in it but will instead be done in a Japanese style with nuki (penetrating ties), I am proceeding with the French layout method of using the footprints of the parts on plan as the main organizing and sizing factor. More on that in a later post possibly. For now, I will forge on to finish this sawhorse off so I can start using it.

The last step in the main frame are the interior x-braces. After a bit of wasting with the circular saw and some router work, I had defined one-half of each tenon, and for the rest it was saw time, and long rips were the order of the day:

Here I am experiencing a little separation anxiety:

Rough-out complete:

The upper end of the brace has a less-radically configured tenon and barbe:

Rough out done:

A while later I was sitting pretty with all the tenons on the interior braces ready for final trimming and fitting:

From the other side the lower end tenons can be seen more clearly:

So, in the book it shows these tenons passing right through the long-side braces, which I now realize is impossible, assembly-wise. The only way the braces, once lapped to one another, can pass through the long brace member is if they have an abutment on their upper end that is 90˚ to the axis of the long side brace. Just to show you how far away from that these are after rough cut out, I have laid a small metric square on to roughly show that 90˚line on the tenon:

Obviously, the tenon once cut would fall short of coming all the way through, thus I will not be doing through tenons. Here's the lower end, where the situation is even more extreme:

So, I will be trimming the tenon to a length of, oh, 1/2" or 3/4" in length or thereabouts and cutting some blind angled mortises in the long side braces. These tenons will either be left dry or possibly glued at time of final assembly.

That's what's on the slate for tomorrow - mortising and probably some fitting up. I hope you will return then to see how it comes out. Another couple of posts and this thread will be done, believe it or not. Your visit is always appreciated, and comments are most welcome.

--> Go to post XIX


  1. <>

    I think that the use of crosses is also a matter of tradition. If one looks to wooden, sometimes medieval, houses (maison à colombages) one can see that in some styles crosses are massively used and other times not at all. I think it is an expression of local style. Some of these houses can be seriously out of plumb, most likely those without crosses. In an old house nothing is straight. Examples of styles can be seen on:à_colombages

    I think that stabilising roofs is important, as many old roofs have a stability problem. And certainly Mansard roofs where the beams do not rest on massive walls, and the volume of the structure is mainly kept to the roof surface itself.

    Another notable technique, seen on some examples of colombages, to set each floor more outward than the previous, is also about stability, this one to keep the horizontal beams and floors balanced and thus straight, as well as protect them from the weather.


  2. Damien,

    thanks so much for your comment. I am sure that style/tradition, and its later blind imitation, plays a very large role in architecture the world over. My point however was not about the use of bracing and crosses in wall structures, which are obvious to view, but was directed in particular to the use of cross-bracing in splayed post structures, like a hipped roof.

    Your point about stabilizing roofs, in regards to these St. André's Crosses at least, doesn't entirely make sense to me, but is food for thought. Hipped and Mansard roofs are inherently stable shapes after all. Yes, if the foundation subsides or a wall buckles/deforms, then the roof will be subject to unusual loads which will likely lead to its ultimate deformation. There are buildings which have lasted for a long time and have a less-than ideal roof structural design - case in point being the numerous roofs in the area in which I now live that are of the common rafter type, with the ridgeline distinctly sagged in the middle and wall bulging outward pronouncedly at the mid-plate area. Not something worth imitating despite its durability.

    Given that braces work in compression, they will both help and harm the roof as these loads from wall/foundation problems are applied in various directions. However, the issue is not the roof in this scenario, but the foundation and/or walls, and to design the roof to compensate for wall or foundation insufficiencies seems a slightly odd way to go about things. Why not improve foundation and wall frame detailing?

    Gabled roofs are not so stable as hipped ones of course, and the addition of some form of bracing in the roof plane makes good sense in that case.

    I have no issue with wind bracing on plumb walls (in fact I like that idea a lot), providing it is detailed properly.

    Conversely, I wonder how many contemporary French timber structures employ St. A's crosses in the roofs they build?

    In regards to wall bracing it has been noted by other writers that wealth display on old half-timbered structures was accomplished, in one way, by adding lots of extra and decorative bracing to the outside walls. Many of the buildings linked to in your comment feature purely decorative bracing in many places.

    Your final point, about jettying one floor above the next, as far as I know had more to do with working around tax rates which were based upon ground floor building footprints, and for the purpose of providing some overhang to protect merchant wares or operations (and the lower wall framing, as you noted) at ground level, as the roofs in general in Western European framing have but modest overhangs. I don't think that an argument can be made for the jettying of floors adding especially to overall stability, as the structures become top-heavy if it is carried on too much. It's more logical and easier to do when buildings are located one against another in a row. The most *stable* arrangement for a detached structure, generally speaking, is to have a wide base and a smaller, lighter top structure, like a pyramid. Of course, a pyramid would not be a practical residential building form!


  3. Hi Chris,

    Seems you've mislead us as well - wasn't last entry #26? At least its not a big deal to fix and wouldn't matter much if never fixed (its not a tax form after all).
    Anyway, enjoying the story.


  4. Steve,

    jeez, again! Thanks for noticing, and duly corrected.

    And one more point in regards to Damien's comments, as I've had a little time to further reflect:

    The maison à colombages linked to in your post are not such a reasonable point of example for the issues I raised in regards to bracing, as such all-timber structures were already comparatively rare by the late 1800's in much of Europe. From what I gather the forest base was largely destroyed by the mid-1400's in many parts of Western Europe, due to both timber construction (buildings and ships) and, more particularly, the use of timber for fuel in the production of iron. In Mazerolle's book, along with two other period French carpentry books I have, there are no depictions of large buildings like those half-timbered ones in the Wikipedia link - virtually all the roofs to be seen described in the books are parked atop stone or brick walls. The only free-standing all-timber structures described, from this sample at least, are bridge centers, which are temporary structures, or small open air pavilions and the like. It seems by 1850 that the focus of timber carpentry in France, at least, was the roof only. So, the argument about the roof needing to hold together in the case of wall and foundation subsidence is less apt I would think when thick masonry walls are used.


  5. Chris,

    You are much more a specialist than I am. But I am also interested in the why. I agree that modern day carpentry lacks those complex diagonals, they look more like a torsion box with a large number of short transversal pieces. But apparently older Mansart roofs were different. I can imagine that a carpenter working on an early Mansart, say in 1540 on the Louvre, can have made repairs on a Gothic roof like that of Notre-Dame some years earlier, or the renovations of any second house in town. I expect the Notre-Dame roof to be a big alignment of A like structures crossed by a X. The X is in a different direction but heavily present. My impression is that a Mansart roof can be seen as a Gothic roof where the top part is removed, and the remaining lower structure stabilised by internal walls or added structures. For those carpenters Gothic or Mansart could have been business as usual, same skills, similar techniques.

    About jettying floors. We have (?had) taxes for extruding elements on a facade. A balcony is OK but you have to pay more taxes. A closed balcony? Even better, very impressive, I see you are a wealthy person. Etc. There is no way to avoid taxes. But as well as sagged roofs, I have seen many old sagged floors in masonry houses. Putting the weight of the wooden façade as counterweight on the extremities of a beam can help to reduce that problem. Of course you need to compensate the small foot-base with a stronger structure. We have the extra problem that beams had to be fixed on the front and back walls, and not to the common side walls. For narrow town houses this means that the beams are set to the greatest length.

    Apart from all that, I am quite happy and impressed to see your tréteau taking form.



All comments are moderated, so if you're planning to spam this, know now that your clicking and pasting is in vain. I do read the comments before posting, so your mission is doomed from the outset. All this time and effort trying to put your inane spam onto blogs -- is this how you want to spend your time on earth?

Please do me the courtesy of appending your name to your comment, even if posting under the 'anonymous' option. No name = deleted.

Comments NOT accepted include:

-those containing links unrelated to blog content
-spam of any kind, or ham for that matter
-did I mention that attempted spam postings will be non-starters?