Tuesday, August 10, 2010

This One Rings a Bell (5)

Today is a post about progress, the sort one makes by taking a step back and then two forward. Yes, another wormhole was discovered, and I surely fell on into it. I was trapped in there for a good while today.

Here's the problem - a little matter that all started with a couple of lines that weren't quite matched up:


You might be wondering what place in particular I am referring to, so I'll zoom in a little closer to the left-hand fan rafter:


Do you see it now? No? Well, how about we look a little closer yet:


Stands out like a sore thumb, no? Okay, okay, I'll move in closer yet, towards the spot where the middle horizontal line intersects the vertical line on the left side:


If you click on the picture, you'll see that just a hair to the side of the solid line is a little dashed line. Trouble is, that dashed line was the reference, and the solid line snapped slightly over to the side for some reason. A solid line I based a whole bunch of stuff on....

Damn that SketchUp - I shake my tiny fist...

That little kafuffle with the line had effects when I started placing one of the developed fan rafters into place on one of the purlins and things just weren't quite adding up. That's how I found it - tracking backward until I located the discrepancy. Unfortunately, in the end I realized that the repercussions of that line being just a hair wrong were going to have numerous down-stream effects, and more importantly, it meant that I had to re-draw all of the curvilinear parts. -gulp-!

Here's the thing - while the drawing will look absolutely fine upon inspection, unless you magnify in, way in, continuing on with the drawing when it is based on an error, no matter how slight, means I cannot rely upon the data points I have acquired so far in the drawing process. You see, once the drawings are completed, I will be spending quite a while re-drawing key portions of the 2D so as to make a whole bunch of templates with which to shape parts and check alignments. The 3D allows me to be absolutely sure of my results, but it absolutely punishes me if I don't have the 2D portion just perfect. Nothing painful mind you - I imagine some future version of the software might be configured to give the user little jolt of they make an error, who knows. Just kidding, I think...

By getting the 2D perfect, I can rely upon the drawing directly when I draw the templates in full scale, and I will be able to take off various heights and measures from the virtual into the actual with total confidence. So, it's worth taking on the error and fixing it now. That's my stand at least.

Questions of deciding whether to fix an error once discovered or to carry on and either build on top of it or try to compensate for it some way come up all the time in construction. It's all about the best laid plans of mice and men, and this mouse is trying hard to get the plans laid as right as he can.

So, 10 hours later at the drawing board, and I am only just managing to stave off going completely 'round the bend, but here we have it, the Mark II:


Yes, that back step was a royal pain in the you know what, but it paid truly handsome dividends in the end. You see, that little line problem wasn't the only discrepancy haunting me.

Though I had completed the hip rafter layout to the point where the fit between components was decent at the exterior intersection...


... in reality all was not totally fine as there were some problems with the peak of the backing cut on the hip:


That little gap above the hip arris where the kaya-oi meet on their lower surface had been bugging me, and I had decided after that previous trip down the wormhole that I would, you know, how do they say it? Oh, yes: 'deal with it later'.

Trouble was, I hadn't really found a solution for determining the backing cut on a gradually curving hip rafter with a curving purlin meeting it. I have virtually every Japanese layout book there is, and, strangely I might add, none of them seem to cover this matter of the backing on a curved hip rafter with curved kaya-oi. There is one illustration in one of Togashi-sensei's books that shows the hip nose layout including the backing cut, and I had followed it faithfully, several times over in fact, and the results were plain enough: it didn't come together right when the parts were assembled in 3D. The drawing above is proof of that, a product of that method, and when I drew the new hip rafter and kaya-oi, I had the same problem again. So much for the aforementioned 'later'.

I have other roof carpentry layout books in French, German, English, and even Yankee, but the topic of curved hips is really not covered in any semblance of detail. The few odd methods I have come across in those texts I have tried but they don't work when the eave purlin which lies atop the hip is also curving up. They only work for curved hips with with straight parts laying on top of them. A curved hip with flat strips or boards laid upon it is no big deal, but if the eave purlin, kaya-oi, is curved, the backing cut is a tough nut to crack. It has stymied me for a while now.

Well, was a tough nut to crack, for I managed to solve the layout on that today! By determining the correct method, through some reverse-engineering and a sudden flash of insight, I was able to also determine what exactly was wrong in the illustration in Togashi's book. That, my friends is the very first error I have ever discovered in the master's work. While I derive little satisfaction in finding his little mistake, I am tremendously pleased that I have at last solved a vexing puzzle. Maybe I'll write him - not sure if he is still alive though.

In the end, I had to construct new hip rafters, kaya-oi, and komai. So far I have placed the new hips (sumi-ki) and kaya-oi on the roof, and things are fitting just as they should:


I also had picked up a subtle little tip that I noticed discretely marked in one illustration of Togashi's book and from there re-sloped the hip nose cut and re-shaped the hip end slightly:


These slight changes decrease the amount of end grain of the hip sticking out, always a good thing to do.

Here's a look down that hip rafter, showing the curving and slightly twisting backing cut:


Well another adventure in layout-land my friends. Next I hope to take a few more steps forward.

Thanks for your visit and comments always welcome. --> on to post 6

11 comments:

  1. Chris,

    My head is swimming! It is SO complicated and SO fascinating.

    Tom

    ReplyDelete
  2. Hi Tom,

    yes, swimming, with occasional near-drowning episodes. I'm glad to see that you find it fascinating. This stuff keeps me up at night - one of the things that makes it such an excellent path to follow.

    ~Chris

    ReplyDelete
  3. Hi Chris,

    Sorry to hear about the class getting canceled but I certainly know how that goes, had to cancel my last two because of lack of interest.
    The build is just amazing.... I always admire how you just keep at it until you find the answer when there is no where to look for help, as with the saw horse and now the roof for this project.
    Know that even when I don't write anything for a long time I'm always looking.
    Carry on...... Charlie

    ReplyDelete
  4. Hi there, Chris - we're very impressed with your persistance. It is likely the masters never had a computer program to help, so it would be interesting to know how much one of these helps or hinders the designing process.
    Try and get some sleep.
    Love Mum + Mitch.

    ReplyDelete
  5. Well, hello Mum! Nice to see your comment, and to know that you are checking out my blog. Welcome!

    I will say at the outset that many of those who do not use a computer to draw do not understand the way that a computer drawing program works. Or, at least don not understand the way in which I am choosing to use the current software I choose, SketchUp. So it is likely that some, at least, will have formed an erroneous impression about computer aided drawing and what it entails.

    First of all, using the computer is not some matter of having it do 'all the work' and it telling me somehow whether things connect or do not connect. It's not quite like magic. I can't simply tell the computer to build me a fan rafter roof of such-and-such dimensions and, shazam!, it's done. It takes me nearly as long on the computer to draw something as I do with traditional drafting tools on paper. But there are several advantages.


    I draw in an identical fashion on the computer as I would in a shop - save for the different tools involved. The computer does make it quick and easy to draw parallel and perpendicular lines, to determine angles, create polygons, and, significantly, to instantly erase mistakes and re-configure things as needed without making a mess of it. I can also toggle off things I don't want to see for a while to give a clearer picture of the portion I am working upon. SketchUp is poor when it comes to drawing circular figures, but other drawing programs are better in that area.

    Now, without the computer, I would do the same 2D drawings as I am doing on the computer now, likely in full scale, and if I was a bit uncertain about how the actual piece would come out from those drawings I might choose to build an actual scale model.

    In SketchUp, I accomplish the same thing, except I virtually model 3D shapes right of the 2D portion of the drawing. It saves using wood for a model, and is completely precise. The downside is that those models can be things of beauty in and of themselves, and in some cases, are valuable historical artifacts.

    The error I mentioned in the above blog post might seem really insignificant, but you know, when I re-did the drawing, that little difference in the line position resulted in the eave fascia curve being 2~3mm raised up at the end of it's curve. That's a significant difference that would be obvious in real wood.

    And by discovering the tiny errors so clearly, I can find the places in the drawing where I am making the mistakes and correct them. Thus I learn a lot in the process. The computer can allow me to effectively 'reverse-engineer' parts as well to try and determine the layout for those things which stump me. Very helpful.

    Finally, as a tool for effectively communicating with clients, 3D drawing is miles ahead of plain 2D elevation and plan views, and unless one is skilled at making perspective drawings on paper, the 2D renderings are all the recourse one has available. With 3D, one can orbit around and look at a piece from any angle or perspective, and this information enables one to make all sorts of design alterations as a result. It's a virtual substitute for doing models or mock-ups, as are common in furniture-making and architecture. I can change woods/finishes/colors/roofing at a click of the mouse, while to re-do a colored pencil sketch on paper in different materials would be a lot of extra work.

    I think that computer drawing has the potential to lead to more sublime designs, but the jury is out on whether a flowering of design will be the ultimate result, or a narrowing. Generally, the driving factor in design, in this culture, is economy.


    In the end, when it comes time to work on the wood, I can apply the same 2D drawing techniques to make templates and shape the piece from there, secure in the knowledge that the layout is correct from the computer drawing work already done.

    I look forward to our next conversation.

    ~Chris xxo

    ReplyDelete
  6. Hi Mum, nice to see you dropping by to say hello.

    I just wrote a long answer but the computer, in some irony to the topic, ate it upon posting. I think I'll take up the question you ask in a blog post shortly as I don't have the energy right now to type that again...

    I look forward to talking soon.

    Chris xxo

    ReplyDelete
  7. Well, that was weird. It looked like it ate my long reply and I lost the whole thing, but now I see it is back. Go figure!

    ~Chris

    ReplyDelete
  8. Are you planning on having a series of ascending purlins? If so, why not just do the backing cut on the hip only at the loci where it receives the kaya-oi and the purlins.

    Ian Lawford

    ReplyDelete
  9. Hi Ian,

    well, though there are ascending purlins, since all the komai as well as the under-eave boards also terminate at the hip rafter, there needs to be a backing cut in those sections as well. It would be more awkward to mortise and dado for those parts on the hip than to cut the backing. If you read ahead in the series you will see that I have made further modifications to the hip section from where it stands at this post in the process.

    ~Chris

    ReplyDelete
  10. Im feeling your frustration on this one....oh the wall doesn't connect to the foundations...thanks autodesk revit...lol

    ReplyDelete
  11. Rebecca,

    yes, CAD can be wonderful and then it can be damnable. Thanks for your comment - glad you can relate.

    ~C

    ReplyDelete

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?