In my previous post on the hip-mounted dormer, I mentioned some issues I had with the interface between the lower end of the noulet, the top of the dormer wall plate, and the adjoining sloped brace and tie which carry the load of the hip. Following the drawing, I initially obtained a hole in the roof, and this led to a solution where I kicked the bottom of the valley piece over to the side a bit more at its lower end. This got rid of the hole, and seemed all in all a tidy solution. I finished off my description of that issue with the following:
"That's my solution and I'm sticking to it until someone can show me some way that the text's proposed version can work cleanly."
Well, I have in fact come to see the issue a bit differently, and have come to consider another way to solve that problem, though, like the solution I had already come up with, not in a way in which the text proposes or indicates. I had some emailing with Tim Moore, a computer graphics guy with a blog on stereotomy (see the link in sidebar to the right of the page), and he suggested a move involving changing the shape of the horizontal tie of the hip support bracing. I looked at his idea in some detail, and after an initial sketch of the changes i felt that while it presented a workable solution, it seemed to wander ever farther from what was shown in the text.
However, further reflection upon my solution ensued. I noticed that while my solution was nice and tidy, it did in fact appear to create a problem in terms of the intersection of the dormer roof planes with the main roof planes. So, I decided to explore in the direction Tim was proposing. After a bit of fiddling, and a cascade of downstream effects which occur from moving a piece here and there, I was able to produce an intersection between parts which kept the lower end of the valley piece aligned to the outer edge of the hip bracing diagonal member.
Here's how that solution looks, and no, it doesn't look a whole lot different than what i came up with previously:
The lower end of the valley is aligned to the outside face of the sloping hip brace, and there are no holes.
With a few parts removed, the slight change effected in the horizontal portion of the hip support assembly is clearer to see -see how it is raised above the dormer wall plate?:
And one more, with all the parts back in place:
So with all that done, I seemed to have found a solution which, though at variance with what was shown in the book in terms of how the central horizontal portion of the hip support truss (some would call it a 'tie' however that term is misleading in this case) was configured.
I then decided to compare the roof surfaces after placing some temporary roof planes on the dormer and immediate surrounding area on the main roof. I then learned that my apprehensions about there being an issue with how the intersection of roof planes would look in regards to my original solution were unfounded.
Here's one solution:
And here's the other:
Not much to pick apart is there? The one with the red plane on the main roof is the revised framing solution described in today's post, while the one with the orange main roof plane is the one I wrote about previously.
So, there is more than one way to solve it, and I'm not sure in the end which is to be preferred. One thing is true, and that is a faithful duplication of the text will result in a hole in the roof. The solutions involve either modifying the central horizontal member of the hip support truss, or swinging the lower end of the valley piece over further. Six of one, and half a dozen of the other I guess.
Thanks for coming by the Carpentry Way. Here's a taste of what I'm currently working on, drawing- and study-wise: