I don't know what to do with this, especially seeing as in the past on the forums there were threads criticizing 'computer' aided design that hinted at the notion that the computer was making the decisions for the designer... a daisy wheel, in this application, is that mindless computer, making decision for the designer, without regard to function, use, materials at hand, solar orientation, etc.

I guess that was my point in questioning DL about his windows.

DL - I think it is great that you plan on making your own windows, but the point is irrelevant as most windows can be had in any shape and size you want from many manufacturers. What I was driving at was the location and size of the windows in relation to use of the building - were heights calibrated to occupants eye level as they might be related to exterior views? Did solar orientation play a role, if relevant, to window location and design? etc. etc. - or did the location and size come about by some geometric convention derived from the use of the wheel (which is what you did)...

As I mentioned - I'm fascinated by all systems of scale / layout / proportion. I'm curious where one draws the line at which decisions are 'blind' to the system, (be it 4x8 sheets, 4x24 panels, 8x8 posts, golden sections, wheel geometry, etc.) and which are made by the designer - in reference and respect to all those other factors that may be important (or not).

A set of dividers is about as close as you can get to a democratic tool. Anyone can use it, it comes with a low cost or even free and or you can make your own. This is what I find attractive about the dividers.

I can not say that for the computer.

Tim

bmike, the placement of the windows was a result of the wheel, but it's not as if the wheel gives me an option of 2 or 3 heights to place the window. There are probably hundreds of different places I could put the thing by using intersection points, I chose these paricular location because it was aesthetically pleasing, and because that particular location seemed reasonable. The width and height of the windows was not laid out according to the 'master wheel' at all, but rather it was made to fit the space between the posts and still create the desired proportions. If you look closely at the drawing, you can see each window is slightly overlaid with a star of David, that is how I determined its proportions. I made a new circle for each window. In reality when doing this on the completed structure, the doors and windows will be set inside of frames to keep them at consistent heights and widths, so as to avoid the irregularity of the posts. You are right that in most projects all of this would be unnecessary, but like I said it's just something I wanted to do in keeping with the medieval theme of the design. Somehow having vinyl clad Andersen Windows or some such would just be out of place, although certainly a whole lot easier!

The wheel as I use it really is not as commanding as it may seem. It's no different than if you would use graph or grid paper to draw a building design. There you are stuck with the lines of the graph, so how is being stuck with the literally thousands of lines that the wheel can give me any worse or more blind? Look at my layout drawings with the wheel and its lines in the background.

By the way, I foresee the question arising of if you have these supposed thousands of possibilities, how is geometry any better, or how are you staying geometric, or some such. The answer to that is quite simple, some lines are better than others! The best lines of the wheel, what I call 'Category 1', are those lines that are simply connectors between circle overlap points -the lines you see drawn on my basic diagram- especially those that connect opposing or reflected points. These lines are the ones I like to use for the basis of my designs "Category 2" lines are still good, these are the lines made by intersection of 'category 1' lines. These are what I like to use, for example, to establish roof lines. 'category 3'lines are those that are formed by connecting intersections of category 2 lines, or intersections of category 1 lines and circles. Category 4 lines are formed by the intersection of building lines (when they don't fit into a previous category) and any other category 3 or higher line. And the list goes on and on... It sounds complicated but it's really all quite natural and fairly obvious. That's the system I use, which I think yields a good combination of limitation (because limitation yields strength) and freedom.

And that last point is an important point Limitation yields Strength and I could also say limitation yields beauty

Tim - perhaps you should have been more clear on this distinction in previous posts. It would have changed the conversation.


DL - Yes, there is power within limits and a fascinating book of similar name. Thanks for the thoughtful replies, it has shed some light on how you are using the wheel. And - its a cool study - my questions were intended to sort out the why's of how you were getting your results.

Any system we use to design a building is a set of limits, and from those limits a 'normal' arises. Normal, after all, is a relative term for something that conforms to the standard limits in a given system. A 'normal' door, for example, would conform to the standard industrial sizes of 2'8, 3'0, etc. anything other than that would be unusual

So we could assume that the Daisy Wheel creates its own normal that may or may not conform to the imperial or metric systems' normal. But I suspect that we could easily get things *close enough* to escape the building inspector's scrutiny. My building, for example, is close enough that I could call it a 24x14 and not have any problems with the system, even though it is about 1 1/2 inches off from that. The irrational numbers that the wheel is going to inevitably give you can easily be represented as real numbers for the sake of those of us that do not think geometrically, even if we step them off somehow when we actually lay out and join the timbers (whether it be with dividers or rods, or whatever system you choose to use). That is the reason that I have referred to my building as 24x14, even though that is a little bit inaccurate (a hair more than 1 1/2 inches off)

I was thinking of the problem of glass for my windows, for example. What am I going to do about that, I can't just send out to glass guy and ask for a piece cut the square root of some un-square-rootable number, or give him a set of dividers and ask him to come up with the glass size for me geometrically? But it would work just fine if I were to give him a size measured to fractions of an inch. What's important is that the frame for the window glass, which I will make myself, line up geometrically. It doesn't matter if the glass has a 1/8 inch of room or so inside the frame, where only bugs and dust and architectural historians a hundred years from now are going to be able to tell that I 'cheated'.

http://www.traditionaltimberframe.com/V1_0/index.php?mod=frenchscribe&ac=procedure

For all your scribing needs, this describes the process in detail.

After having researched it, It seems that the French Scribe, or at least the alignment of the timbers to a full-scale ground plan (which is in fact only part of the whole french scribe process) is in fact a method used across Europe. I have seen many pictures of German carpenters laying out buildings along full-scale ground layouts.

I have also discovered a solution to layout on irregular ground. The method shown in that link is designed to be used on a flat shop floor, however that is not the way it is always done. Historically the layout would often be created exactly where the building was to be raised. The solution I have seen used is to take boards and lay them out on the ground with chalk lines down the middle, squaring the lines using the Pythagorean Theorem instead of using a giant compass. The layout can then be made with lines snapped on a series of boards. The advantage to this is that it is easier and more accurate to step off measurements on flat boards than on uneven ground, and with this method you can lay out the structure without needing to have a shop floor that big.

This all reminds me, at the base of the great pyramid in Egypt a series of holes were once discovered that, when connected, supposedly create an exact elevation of the pyramid, complete with the details of the internal chambers and all that. In other words, the exact geometry of this enigma was established with a plan, and the plan seems to suggest that the multiple chambers were not, in fact, the result of a Pharaoh who kept changing his mind. It would appear that this method of scribe layout is incredibly ancient. I wonder if such a system was used by the ancient Greeks to build their amazing temples and amphitheaters

I created a full scale layout today of my building designed with the daisy wheel, the primary reason for this was to test methods and concept, and to have some practice doing such a layout when it comes time to make the building (I would hope to be able to get started by this fall, but don't realistically expect that to happen til the middle of next summer)

My final layout is going to be made right on the ground where the building is to be erected, partially so that this personal project does not occupy valuable floor space, and also in keeping with the medieval spirit of this project -I have read that all buildings built prior to the industrial revolution were done like this, lade out and cut right there on sight, and that shop layout is a modern development.

I started out by using boards with lines chalked on them to establish a right angle. I did this by using the Pythagorean Theorem. I measured down 1 line 3 units, down another 4 units, and then had a third board with 5 units marked off on it, then manipulated all 3 boards until the 3 segments matched up to form a perfect 3-4-5 right triangle, thus establishing the right angle.


next I established the rectangle of the floor plan with more boards, although in hindsight the boards were necessary only for establishing right angles, and were not needed after that point.


While creating the layout, I made a series of Rods based on dimensions on the daisy wheel diagram. These rods were of various proportions, with their proportions clearly marked to avoid confusion. They were at least 4 times the scale of the drawing and some used for smaller measurements were the full 24 times the drawing lines. These consist of pieces of wood with 2 20d spikes driven through at exact points determined by the dividers, and the points of these then carefully aligned to ensure they too are the the same exact distance, Any error here would only compound over time. The spikes were driven all the way through, so that they would be long enough to stick into the ground so that I would not need boards everywhere. The ground at this particular spot is flat enough to do that and still be accurate, and measurements taken afterward confirm the accuracy of this method.


I proceeded to lay out all important reference lines, using my trammel point compasses to step off various measurements, all the while discovering how frequently and beautifully things repeat and relate to each other. If I had known this, I could have gotten by with maybe half as many rods!

View of roof geometry layout

closeup of roof geometry

pitch change, plate, post and tie

right click on pictures and click 'view image' to see a larger view, details don't show up very well otherwise.

more pictures at:
http://s538.photobucket.com/albums/ff345/HiddenOrder/Timber%20Framing/

I made 1 or two minor adjustments to some of my measurements when I switched methods for stepping them off. I decided to see how well I would fare by not measuring 1 end of the floor plan but trusting in the accuracy of my right angles and lines, just to see how well the system worked, the result being that it was accurate to within 1/2" in 24' of length, which is very good! At one point I had to step off to find the center point of the end walls, and I was very pleased at how accurate it came out having not measured 1 wall at all. (I did then adjust that measurement so that it would be perfect)

I was very pleased at how accurately I could make right angles using this method, and was especially pleased that I was able to complete the entire layout to very exacting standards of accuracy completely by myself. Needless to say, I am very excited to begin laying out the joinery!

I am very satisfied with these results, and for me at least it proves the concept of this method of daisy wheel layout. The full scale layout allows you to see things that aren't immediately evident on paper, and the laying out process gives you a much better understanding of the proportions involved.

It also revealed another thing to me. The rulers that I used to set my dividers are supposedly accurate to within 0.005" or maybe it 0.0005", whatever it is it's a very tight tolerance. My good old Stanley measuring tape, however, does not reflect this same degree of accuracy. Having set my dividers to 4", the slight almost unnoticeable difference between the two comes out to a difference of a little over 1" along the 24 foot long wall. So unfortunately I won't be able to measure proportions and relationships as easily as I had hoped.

Correction, the long wall was 24 5/8" long according to Stanley, so the difference between measuring devices comes out to 5/8"

Hi DL, please keep us posted on your progress... your threads are some of the more exciting ones on this forum

re windows... I designed one of our feature windows with the wheel, the window co. simply asked for a template and are building it based on it vs. measurments. The guy was amazed to see the wheel still in use, we had a good long conversation about it...

Perhaps a title for this post could be Moving Beyond the Wheel

In studying this system and working out the various aspects of my design, I think that there is much more in the realm of circular geometry than what the wheel itself shows you, at least right off.

I remember once having read a description of how Canterbury cathedral was laid out according to the system of Ad Triangulum. This, as I understand it, is a system of laying out rectangles by means of squares. This cathedral was limited by the narrow frontage of the lot it was built on, and so this measurement was used to derive all other measurements, down to the placement of the altar within the chapel. Essentially this was done by taking that measurement and various sections of it and drawing a series of circles on the ground to create the plan. Below are examples of the Euclidean Geometry of Canterbury Cathedral

The Geometry of Lanfranc’s Cathedral 1070-1096/1120

The Geometry of the Choir of 1120