Architectural Details and Ornaments

[Laurence Finston]

This is an idea I've been carrying around with me for a long time:

I'd like to make plaster models of architectural details and ornaments, for their own sake and also to be able to use them for embossing paper. I haven't yet tried to print from plaster but I don't see why it wouldn't work. If it does, I plan on doing that, too.

The embossing part is why I'm posting this under "Design Threads" rather than "Other Things We Do & Make": The results (assuming there are any) will be models and made of paper, though admittedly not the typical kind of paper model. Unfortunately, it doesn't seem practical to me to make these things in a size that would fit with architectural models in a scale that's normally used. Certainly not in the prototype stage.

The first image is page 207 of Owen Jones' 19th century book "Plans, elevations, sections, and details of the Alhambra" (public domain, available at the Internet Archive). I'm particularly interested in the ornament from the Alhambra and in Islamic ornamentation in general. (M.C. Escher was also interested in these things and he's one of my favorite artists.)

The second image is the output from a MetaPost program I'm working on to duplicate as closely as possible the design from the first image in the form of a technical drawing. I've solved the basic problem, the rest is mostly repetition. The source code is available here:

MetaPost code: alhambra_207.mp\src - 3dldf.git - 3DLDF
TeX code: alhambra_207.txt\src - 3dldf.git - 3DLDF

As is obvious from the MetaPost drawing (but not quite so obvious from the original), the pattern is based on hexagons. They are slightly disguised by the wavy lines making up their edges. One of the nice things about MetaPost is that, given a set of points, it will try to generate the most "pleasing" curve through them, whereby the user can give MetaPost hints about how the curve should be constructed. In order to make the wavy lines, I didn't have to use the sine function (the most obvious way of doing this); instead, all I had to do was indicate that the direction of the curve should be horizontal when entering and leaving the two peaks of each curve ("dir{0}").

The next three images are details from buildings in Göttingen, Germany, where I live, and the last is a baluster from the (demolished) old Chicago Stock Exchange Building. According to Wiktionary, a baluster is "[a] short column used in a group to support a rail, as commonly found on the side of a stairway; a banister." I'd never heard the term before. These are the sorts of things I have in mind.

The Stock Exchange building was designed by Adler and Sullivan and I'm quite interested in Louis Sullivan's architecture. I originally thought of possibly doing an ornament from the Carson, Pirie, Scott building in downtown Chicago, but they are probably too elaborate. I used to be in downtown Chicago regularly for school or work and always liked the very fancy ornamentation on that building.

[Siwi]

This sounds like an interesting idea. The alternative to plaster would be 3D printing, which might be more durable than plaster if you intend on rubbing tools over it to emboss the paper. I would be curious to know what method you envisage to make sure the paper/card holds the contours of the design and could if needed be coloured to match the original. It would seem that any design would need to take into account the way the paper might deform.

[Laurence Finston]

Originally Posted by Siwi

This sounds like an interesting idea. The alternative to plaster would be 3D printing, which might be more durable than plaster if you intend on rubbing tools over it to emboss the paper. I would be curious to know what method you envisage to make sure the paper/card holds the contours of the design and could if needed be coloured to match the original. It would seem that any design would need to take into account the way the paper might deform.

Thank you for your response and interest. The university library here has 3D printers which I could use. I haven't tried this yet. MetaPost can produce SVG files directly, which is what I would need. I'll keep this possibility in mind.

I think plaster would be durable enough for rubbing. The alabaster plaster I've been using seems plenty hard. However, even if it turns out that it's not, I can always cast a new form. I plan on having both positives and negatives made of plaster that are only used for making more molds, so they should last a long time.

In addition, I wasn't planning to try to use them for embossing dry paper. I think for that I would need a steel or brass stamp, like the notary public's seal my Dad used to have. Instead, I plan to try soaking 100% rag watercolor paper and pressing it (gently) into the mold and also papier-mâché. As I posted on another thread, I was able to get dried papier-mâché out of an alabaster plaster mold, whereas I wasn't able to do this with a mold made of ordinary modelling plaster.

Coloring the watercolor paper or the papier-mâché shouldn't be a problem. For the former I would use watercolor paint, which soaks in and binds with the paper and for the later I could use tempera, either wet or powdered. It would be a waste of expensive material to use watercolor paint for tinting papier-mâché.

The deformation is a good point. Papier-mâché will shrink. I don't know how the watercolor will behave. I would expect that it would swell from the soaking and return to more-or-less its original dimensions when it dries. I think I would have to measure the results and try to determine the degree of shrinkage, which would probably differ depending on the specific paper I use.

[Laurence Finston]

I was working on the technical drawing in the first image. Then I realized that I don't need the colors for making a model out of clay, wax or plaster, so I made a black-and-white version, which was much easier (second image). I've also attached the PDF file.

I put a piece of watercolor paper in to soak in distilled water. I figure it will take a couple of days before it's absorbed enough, so by the time I've finished a form it should be ready .

[Laurence Finston]

I've made a new version of the PDF file with a section of the pattern clipped to a rectangle which is approximately the largest one that will fit into a 26cm spring form with a small margin. In fact, it's the outer diameter of the form at the top that's 26cm and the walls have a thickness of a couple of millimeters. In addition, they taper so that the effective diameter at the bottom is almost exactly 24cm. I'd already determined that only the bottom of a cast is practically usable because the top won't be completely smooth and flat. This is especially so with wax, which dishes, but even with plaster the top may have irregularities. In my experience, it's just not worth it to try to smooth and flatten the top of a cast; it's better to just use a bigger form.

There's a non-reversed and a reversed version and I've added crosshairs and guidelines and this time I remembered to draw squares around the crosshairs for cutting out.

The next step is to cut out stencils. I'll need two, one for the hexagons and stars and the other for the triangles with the wavy edges. I'm using 250 g/m2 cardstock this time. It's more work to cut out than with the 160 g/m2 stock but the result is that much more robust so I think it's worth it.

I'm leaning toward transferring the drawing directly to plaster and carving it because I think I could get much sharper results than with wax or clay. Another reason for not using clay is that I don't want to ruin the stencils, seeing as how it will be a lot of work cutting them out. Of course, I could transfer the drawing to a slab of clay by pricking through a print-out, but that wouldn't be as accurate as using a stencil.

To make the JPEG versions of the drawings, I used `pdfseparate' to separate the pages from the rest of the document. Then I loaded them into GIMP and exported them as JPEG files, using the default settings. I'm happy with the results. I used to use Imagemagick's `convert' to try to do this but that didn't work well.

[Mike Stamper]

In the UK this type of pattern making is called Pargetting.

Type it into Google then select images - you will find lots of patterns.

Here's another process - Repousse. This generally refers to making patters in metal sheets using a hammer. I've just tried making patters by putting a sheet of paper on a thick soft card then drawing with a blunt pair of tweezers I also used the top end of a paint brush (the end with out (bristles)

I hope this helps

Mike

[Laurence Finston]

Originally Posted by Mike Stamper

In the UK this type of pattern making is called Pargetting.

Thanks for the information. I used to listen to "The Archers", so I knew the name "Pargetter", but I never thought about what it might mean.

According to Wikipedia, "The term, if not the practice, is particularly associated with the English counties of Suffolk and Essex", so it's not surprising that I wouldn't have heard it.

Here's another process - Repousse. This generally refers to making patters in metal sheets using a hammer. I've just tried making patters by putting a sheet of paper on a thick soft card then drawing with a blunt pair of tweezers I also used the top end of a paint brush (the end with out (bristles)

I'd heard the term "repoussé" but it wouldn't have occurred to me. I'd like to try this but at present I'm limited to techniques that don't make too much noise on account of my neighbours, so lots of hammering is definitely out.

However, in Germany, there's a craft that's fairly popular called "Metalldrücken" (metal pressing). It uses foil or very thin sheet metal that you press with tools like the ones in the photos. You can press onto a felt map, like the one shown, or use foam rubber sheets, like the ones sold at craft and hobby stores ("Moosgummi" in German). You can also fold it. Sometimes, you see ashtrays made this way.

The only thing I've done so far with this technique is the letters, which I've photographed along with some I made with felt. I stitched them onto photo board using embroidery thread or, in the case of the "W" with the sequins, ordinary sewing thread. I think I must have done this around 2005. I've recently dug out my materials for this technique and would like to give it a real try sometime.

Another thing I'd be interested in trying is brass rubbing. When I visited England with my parents in the mid-1970s, that seemed to be pretty popular. At some famous churchyard or other, you could buy brass rubbing materials and you could see people doing it in broad daylight. In Germany, it seems to be completely unknown. I have never seen it being done here, anywhere.

[Laurence Finston]

I've cut out the first stencil. It was tedious but not as bad as I'd expected. This degree of complexity is pretty close to my limit for what I would want to do by hand. I will definitely consider getting a cutting plotter. I think it's good to know how to do things by hand but I'm glad I have a circular saw, for example, and don't have to saw through planks with a hand saw.

I've decided to make 4 stencils, this one, two for the diagonal wavy lines and one for the horizontal ones. I could probably get by with 3 but I think 4 would be better. I might even make an additional one for the surrounding rectangle.

This paper dulls the X-acto blade quickly, so I wore one out with not much more than just this one stencil.

[Laurence Finston]

I've updated the plans for the stencils. I've added guidelines for cutting out the whole stencils. My printer doesn't print the outer c. .25cm and I want to use the partial circle to align with the plaster (or wax) disk, so the rectangle for cutting out is somewhat within the top and bottom margins. I also added a crosshair at the center of the pattern. Finally, I added plans for a stencil just for the surrounding rectangle.

Placing the images horizontally is a little difficult. I'd like to have them centered but the width of the image includes the parts that were clipped and are therefore not visible. This can be fixed, I just haven't gotten around to it and since I'm cutting off the margins anyway, it isn't really important.

It's also not important for the pattern to be perfectly centered on the disk, but following the principle "neatness counts" I've decided to try to do this. Experience tells me that this actually may turn out to be important, if not now, then for a future application. In other words, I might regret it if I don't do it. However, it's not completely straightforward finding the center of a circle if it's not already known. There are graphical ways of doing this but they seem to me to be too much trouble. There are also tools for this purpose which work according to the same principle: Zentrumsfinder, Zentrierwinkel, Zentrierschablone

Unfortunately, I don't (yet) own one of these. Therefore, I've decided to do this a simpler way, i.e., by cutting a circle of the right size (here, 24cm) out of a piece of Bristol board. Then I will cut a small circle out at the center and at a couple of other positions with crosshairs around them.

I bought the circle cutters in the photos a long time ago (the prices for the cutter on the left and the bar are in deutschmarks) and maybe tested them once or twice. I've never really used them and never opened the package for the extension bar.

[Mike Stamper]

I didn't mean you should use metal !

Just use the technique using paper and card and a blunt stick or similar!!

Cheers
Mike