During this Wildcard week I was free to choose a new “technique” to dive into. As long as I would design and produce something with a digital fabrication process (incorporating computer-aided design and manufacturing) not covered in another assignment. I wish the lab had a knitting or embroidery machine, because that would’ve been a no-brainer choice for me. However, since that wasn’t an option, I was really happy to see that origami was a suggested direction. Since I have a strong interest in origami (multiple lights in our house are origami-based and self-created), that felt like a great technique to explore this week.
My (self-appointed) task for this week is:
- Explore ways to use machines to pre-cut/pre-etch a paper that can then be folded into origami structures, without tearing the paper
Showing some final results of the successful tessellated folds for the
tl;dr and hopefully showing you why the rest of this very long blog could be interesting to read (⌐■_■)
During the “Machine Design” week, I initially wanted to cover the underside of each petal with an origami tessellated structure that would fold in and unfold as the petal opened and closed.
I tried several ways of using machines to pre-create a fold pattern for me, which I could then easily fold into perfect structures. Sadly, I couldn’t get a process to work. Everything resulted in tears in the paper.
Since molding and casting wasn’t quite my thing, and I had zero ideas of what to do with composites, I actually wanted to continue my origami experiments for this Wildcard week. My goal would be to try and have a machine create good fold etchings that I could fold along which wouldn’t tear the paper.
Note | Because I worked my a** off during the “Machine Design” week to get our acrylic folding light operational, I hope that it’s ok that I see my origami experiments from that week as already being part of this Wildcard week. The origami experiments were not incorporated into the folding light at all, so it can be seen as completely separate.
I spend at least 2-3 hours scouring the web, looking for inspiration on what folds I could use. I gathered my favorite visuals in a Pinterest board.
Laser Cutting Tests | 1
I thought that the laser cutter would be a really efficient way to “prepare” paper for the folds of the pattern. The laser would etch out the lines perfectly, making the whole folding process much easier. I wasn’t sure though if the laser could be that subtle on paper, to only etch and not burn through.
As a very first and quick test I drew two straight lines in the laser cutter’s software (LightBurn) and did some tests on a few paper samples found in the general paper drawer.
I started out by setting the power to the minimum possible (below this the laser doesn’t turn on) of 10% (both the max power and corner power), and a speed of 200. This actually made some really nice etches in my paper sample. It was definitely too slow for the outer ends of the line though, burning through the paper, where the laser head wasn’t yet moving fast enough.
There was also a thin sheet of plastic which seemed interesting. I did a few tries (15%+75 was too powerful, 10%+100 was good), but even if the plastic was only lightly etched, once I would fold it, it would tear up (image above right).
Loes then suggested that we’d go visit the Textile lab on the floor above ours to ask for ideas on materials to use, perhaps something that could be folded. There we were introduced to a really gorgeous, milky white, thick-ish paper that they’d use for pattern drawing. It was Cromatico Transwhite and I got a thin stripe of the 140 gms and 180 gms paper that they had.
I put the two Cromatico strips in the laser cutter and did a test on 10%+200 and 10%+300. In both cases the paper was nicely etched and it folded really nicely along that edge.
The tests seemed to have shown that creating origami folds with the laser cutter could indeed be an avenue to really dive into.
Creating Test Folds
After the laser cutting tests on straight lines I wanted to create some of the folds I’d saved to my Pinterest board. I started with another long look through my board, opening up the links of the images that I’d saved to see if there was more useful information, such as the patterns of some folds. I also looked through some papers written about origami architecture, which were really fascinating.
I eventually came across the amazingly helpful Origami Simulator where you can see examples of origami (and more) in 3D. Furthermore, they also share the pattern that you can download as an SVG file.
I couldn’t be at the lab that day (covid rules) and sadly don’t have a laser cutter at home (I wish that was possible!) but I did have a pen plotter that I could use to draw out the fold lines for me.
Within the help files of the Origami Simulator I found that mountain folds are red and valley folds are blue lines. Where the opacity of the line indicates the final fold angle (
180° = 1.0, 90° = 0.5, 0° = 0).
I loaded an SVG pattern into InkScape. I really don’t like InkScape, but my pen plotter, an AxiDraw V3/A3 has a plugin for InkScape which makes it easy to plot something quickly. For my “hobby” pen plots where I create my own sketches, I use a different method and use the AxiDraw CLI, just to be able to bypass InkScape.
With the pattern in InkScape, I combined all the red lines into one layer and all the blue ones into another. Using the AxiDraw plugin, I can draw each layer with the pen plotter separately.
I then grabbed the thinnest pink and blue pens that I have and did two passes on an A3 sheet of paper to draw out the pattern.
That’s where the easy part ended, because now the tedious manual labor started. I first used a knife and ruler to etch out all the drawn lines on the paper (I wanted to mimic what a laser cutter would do).
After that I would have to fold each of the lines. Making sure to only use mountain folds for the red lines and valley folds for the blue lines, which isn’t always easy, because these kinds of tessellations have blue and red lines crossing into each other.
And the final, but by far most difficult step was to actually fold the paper into the pattern. I started with Resch’s triangle tessellation. However, after about an hour of folding did I have to admit that I just wasn’t able to get the folds to stick. I could only get one small hexagon of triangles to fold upward, but I somehow couldn’t expand it beyond that first tiny corner.
Miura Ori Tesselation
I decided that I should try a different pattern and downloaded the Miura Ori (sharp angle) tessellation, had it plotted, etched it out and started folding. This is a much more well-known pattern and I could thankfully find some videos that showed tips on how to approach these folds. It still took a long while, but eventually I got there!
Now that I could hold it in my hands I noticed that it would fold very well along one axis, but was quite stiff along the other. This fold would really only lend itself well to covering a shape that was flat.
I then explored a fold that would be able to fold in two directions; the Waterbomb tessellation. Repeating the steps of downloading the SVG, having my pen plotter print it, etching all the lines with my knife and trying to fold it.
I had a hard time trying to get the folds to come in, and only remembered to watch a video for tips after I was done to see how I should’ve made it easier for myself (*^▽^*)ゞ
I noticed that this design would almost want to become a half sphere.
Resch’s Triangle Tesselation
However, the mostly unfolded Resch’s triangle tesselation paper sheet was still staring at me. I decided to give it one more try. And somehow it seemed to go slightly better this time. Very slowly, but surely I managed to get more triangles to pop-up. And after what seemed ages (about 2 hours?) I had finally folded my ±25x25cm sheet of paper into the pattern! (๑•̀ㅂ•́)ง✧
The paper was definitely showing signs of having been handled a bit brutally at times, but the pattern was clearly showing. And now that the paper had this pattern, it would stay in it. It didn’t contract nearly as much as the Miura Ori or Waterbomb tessellations, but it did contract into every direction.
However, making the folds was still very time consuming, and wouldn’t really scale well. I wondered if it would go better if the etchings were made by the laser cutter; both from the fact that I didn’t have to do the etchings myself, and being able to use sturdier paper that would hopefully pop into the folds more easily.
There was one final pattern that I wanted to try. The simplest of them all, not even origami really. A grid of triangles. I would fold all the lines in both directions, and then see how the paper would react to being squished and pulled apart.
I created a quick triangle pattern design in Adobe Illustrator (a bunch of lines at three different angles) and used a slightly more sturdy paper for this test (which just happened to be black). And although I did actually quite like the visual result, I didn’t quite like how it behaved when compressing it.
I suddenly wondered if that Cromatico paper that we really loved could even handle being folded in two directions as I needed to do for this triangle design. I did a test on the two small trips that I had. But sadly they tore off after doing the reverse fold on the same crease (╥_╥)
Laser Cutting Tests | 2
I started out with two tests one some semi sturdy paper that was already at the Waag. I used Resch’s triangle pattern from Origami Simulator. However, in Adobe Illustrator I had carefully tried to optimize the SVG by joining lines together, grouping lines and placing them in order. All in the hope that this would help with the laser cutter to make as few stops and turns as possible.
However, when I turned on the laser, it completely ignored everything that I’d prepared, and just criss-crossed whichever way it liked across the paper. This sadly meant that it went around corners way too often, slowing down for that and burning through the paper. This paper was almost falling apart when I pulled it out of the machine.
When doing a test out of focus actually didn’t burn through the paper all that much. However, there was a clear difference in how powerful the laser was at the top-left and the bottom-right corner. In the top left it was still etching the paper enough to create folds, but in the lower-right corner of the paper, nothing was etched and there were only some brown spots on the paper.
Henk did some tests and thought that perhaps the laser cutter’s bed wasn’t completely level (anymore). But the process to adjust that was quite difficult I understood? (I was later thinking if it could also be partially due to dirty mirrors along the laser’s path?).
I did investigate if perhaps LightBurn could (basically) create dashed lines: never slowing down along a line, but turning the laser on and off along that path. And there was indeed a way. However, this had to be done in LightBurn itself, where you set the Cut and Skip length in the Cut Settings. And in my fold patterns these lengths would be so specific, and they would have to start at such specific locations, that I felt that was a bridge too far to even try and figure out.
Vinyl Cutter Tests | 1
While discussing the issue with the laser cutter and the corners with Henk and Loes, Henk told me to try the vinyl cutter. That machine wouldn’t “burn” through any corners because it slowed down.
At first I’d dismissed the vinyl cutter, because I felt that it wouldn’t be as precise or as fast as the laser cutter on large sheets of paper. But after I saw that it was probably not going to work the way I needed on the laser cutter, I thought the vinyl cutter could indeed be a viable alternative.
I went to the local paper shop (De Vlieger) to get some of that lovely Cromatico Transwhite paper. While I was there I also asked if there was some other paper that might be interesting for the folding that I had in mind, and they recommended the Pergamenta, and I also bought two sheets of Polyester film mat.
Back at the lab I opened mods to see if there was a way to make mods follow the lines in an SVG directly, and not interpret each line as a little rectangle to go around. For example, for my pen plotter, it doesn’t matter how thick I’ve set a line to be, it will draw one line from the starting point to the end.
See the image below-left as an example where I loaded Resch’s pattern of lines, and mods draws the toolpaths at the outsides of the SVG lines. Making these lines really thin didn’t make a difference. In that case mods wouldn’t quite see the line at all and end up making weird back-and-forth toolpaths along the line (image below-right).
The issue is that mods takes an SVG but turns it into a black-and-white bitmap image to look at the contrast and draw the toolpaths around that result. I instead needed mods to take the “declared paths” from within the SVG directly, and use that for toolpaths.
Thankfully I didn’t have to investigate if I could adjust the default setup for the Roland vinyl cutter in mods to do this, because Henk said that you could also use Adobe Illustrator to send commands to the vinyl cutter.
Together with Michelle I looked into this. We drew a single straight line in Adobe Illustrator and clicked Print (from the File menu). In the Print pop-up window make sure to select the Roland GX-24 as the printer. Next, the paper size needs to be corrected: Click Setup in the lower left corner, in the new window select Preferences, and in the this new window you can set the Cutting Area width and height, which you can read off from the Roland printer itself.
For our first line the vinyl cutter still cut both back and forth along the same line, even though it was quite thin. Michelle then tried setting the line thickness to
0.000001 pt and this worked! The cutter would only go over the line once (^ᗨ^)
The pressure was still too much, so the paper was cut all the way through. Michelle and I couldn’t find a way to set the force within the print window of Illustrator. However, the force can be adjusted on the Roland machine itself. Press the Force button, click the right arrow button once to go into the mode to make adjustments, and then press the up and down arrow keys to adjust the force. It can be set from a minimum of
There is one other way to fine-tune the force, and that is with the Pen force. This is a little slider along the bottom of the Roland machine that you can slide between
2 while the vinyl cutter is working, and it will lower or increase the force from whatever you set as the base with the Force button. So in effect I could lower the force a little below
30gf as well.
I did a test on plain paper with Resch’s pattern. Figuring out that a force of
30gf and a pen force of about
-1 was only lightly carving the paper and not cutting through it. The results seemed very interesting.
Optimizing the Pattern
However, I felt that I needed to optimize the pattern paths; to have the cutter go mostly in straight lines, and to connect every piece of line that was colinear and touched (from the Origami Simulator each edge of a triangle was a separate line). I figured that would be most efficient by programming my own version of the pattern.
The pattern when loaded in Adobe Illustrator and on ±A3 paper size:
Vinyl Cutter Tests | 2
I did two tests on plain paper to figure out the right size of the pattern (you couldn’t quite judge this well enough from looking at the pattern on my screen).
I loaded the Cromatico Transwhite 140gms paper and first did a few tests of a straight line to figure out the correct force to use on this thicker paper. I found that a force of
50gf and a pen force of
1 was folding well. I then let it cut out Resch’s pattern across the paper:
After that was done (it was quite slow going still), I began folding, but the more I folded, the less happy I became. The folds were starting to tear! (⑉⊙ȏ⊙)
Remember how the Cromatico paper wasn’t able to be folded in both direction? Well, that was the issue here. Even though I wasn’t trying to fold the same section in both ways, along some lines mountain and valley folds were lying right next to each other. That mean that while I was trying to create a mountain fold on one section, the section next to it that I’d valley folded before was now stretched backwards too far, making it basically fold into a mountain fold as well. And this ripped the paper open.
Henk suggested to try and not use a knife, but is a blunt tip instead, to only crease the fold lines, but not actually cut through the paper. I knew that wouldn’t work on the Cromatico paper because it would still result in the same issue of the paper not being able to handle mountain and valley folds right next to each other. However, this could indeed be interesting to try on the two other types of paper I’d bought; the Pergamenta and the Polyester film.
The next day Henk flipped the blade in the vinyl cutter, because interestingly the back of the knife is a tiny cone shape. I didn’t do any tests on plain paper, but immediately loaded a sheet of Pergamenta paper. I first had to figure out the right pressure that would create lines that would easily fold along.
I started at
70gf, but that wasn’t nearly enough. I then went to
120gf and eventually to the max of
250gf with a pen force set to the max of
2, but even that wasn’t enough to create lines that would fold nicely sadly.
I therefore loaded the Polyester film paper where the max force did make some better lines. I wasn’t 100% sure it would be enough but I just let it carve out the entire pattern. However, when I started folding it, I noticed that it still wasn’t good enough, it was too easy to get a crooked fold. Furthermore, it just wouldn’t hold a fold! No matter how I tried to get the triangles to come up, it wouldn’t keep. The material was too “elastic” in a way and always went back to being mostly flat.
Since the max force hadn’t been enough to even put somewhat decent fold lines into the Pergamenta, I figured I’d give it one last shot with the blade on this paper. I quickly checked if the paper could handle a double fold, and it actually handled that quite well.
I therefore switched the blunt tip to the blade again. Figured out that a force of
40gf with a pen force of around
1 was giving nicely etched lines in the Pergamenta paper. And then let it etch out Resch pattern (about A3 size) on the paper.
However, after it was done, I couldn’t see the pattern! (◎_◎;) I could feel it, but I just couldn’t see it well enough to know which lines should become mountain or valley folds. So for completely unexpected reasons, even this paper was not willing work with me to create my intended origami.
Nevertheless, I waited until it was sunny the next day and I set next to a window to have maximum light. Together with having folded this pattern almost a dozen times by now, I could just see the fold lines well enough, and knew how to fold each section of it, to slowly folded it into Resch’s triangle pattern. But even though this paper could be folded both ways, it appeared that the stress of this rather stiff paper still became too much to handle Resch’s pattern, and it ripped open in a few different sections ಥ﹏ಥ
Interestingly, because the paper was so stiff, and with the fold lines carved by the vinyl cutter, it was pretty easy to “pop” it into this pattern (once I had made all the prefolds).
Laser Cutting Tests | 3
After some time, thinking about yet other methods to try and get good etchings/pre-cuts in the paper that would result in tear-free, crisp folds, I came up with three more options to try, all of which involved the laser cutter:
- After the laser cutter gets a check-up by Henk, try the out-of-focus technique again, perhaps it will perform evenly across the full area of the paper now
- Have the laser follow the pattern with small dashed lines instead of a clear cut, perhaps having some of the paper not cut at all will help keep the corner points together
- Try engraving the lines, where the laser cutter keeps its speed and doesn’t go around corners
After Henk had inspected the laser and tried to fine-tune it for a better output, I started with the laser doing the same rectangle over and over, each time with different settings, all out-of-focus.
I noticed that if it was already just out of focus, it would not draw the lines, but only burn through the corners. I had to crank up the (normal) power of the laser, before it would show up on the paper at all.
With the laser about 1cm moved upward from its focus, it was still burning through the corners. However, it was a relatively small hole, it didn’t affect the line right outside of the exact corner that much, which did happen when I had done tests in-focus before.
Eventually, I found that using a speed of 300, max power of 25% (corner power always at 10%) was giving possibly interesting results. I switched out my test paper for a new sheet and loaded Resch’s pattern; the pattern version I created myself which already had the optimizations reducing the number of corners the laser would stop at to change direction.
However, I quickly noticed that the laser created very inconsistent results. The farther it moved from the top-left corner, the worse the output. In the lower right section the paper was hardly marked at all. It wasn’t a certain angle, it didn’t have anything to do with “getting up to speed” (otherwise the smaller triangle shapes would all look well marked).
I don’t know if that’s just this laser machine or that when setting the laser out-of-focus its power output is extremely variable across the bed in general?
I did still try folding it, but I couldn’t find the fold lines anymore once I moved towards that lower-right corner.
Note | I discovered that when you pause the laser cutter and make adjustments to the cut settings of the layer that it’s working on, it will not be picked up when you un-pause the laser again.
For the next test I used the Perforation mode of LightBurn, the software for the laser cutter. You give a number for the length of the dash and for the gap (given in millimeters). I hoped that the cut parts of each line would help to make easy fold lines, and that the “gap” sections would make the paper stronger against tearing along the fold lines.
As with the out-of-focus tests, I first made some small rectangles to see which setting seemed to produce nice fold lines. In this case I settled on a speed of 200, max power of 15%, and using a dash/gap of 3mm/2mm.
This seemed to go quite well, and after the pattern had been marked, I cut out the central part of the pattern to try folding it all.
It was very easy to make the folds along all the lines. Although there were a few holes in the corners these didn’t tear further into the paper along the lines. With the full sheet pre-folded it was quite easy to pop the triangles into place and create Resch’s pattern. I think it only took me about 2-3 minutes to pop the whole sheet into the fold once I had prefolded all the lines. Quite a difference from that very first attempt that took me more than an hour.
With this test finally having resulted in a success, I wanted to try it on much sturdier paper. I’d brought a few sheets of art paper, 250 gms Bristol, that I use for my pen plots. Using a small test sheet I found that using a speed of 200, max power of 30% (keeping with 3mm dash/2mm gap) was resulting in good foldable lines.
The art paper was A3 in size, which was smaller than the size of the pattern (a few triangles stick out). I therefore couldn’t use weights to hold down the corners, since these could be lasered. Instead I taped the paper to the laser bed using painters tape:
It was a bit more difficult to create the fold lines, with the paper being sturdier. Still, I never misfolded a line (i.e. where the fold goes somewhat next to the intended line). As I expected, with this sturdy paper it was even easier to pop in Resch’s pattern once all the lines had been folded. I only had to pop out some corners and kind of squish the paper inward from the sides. It only took about a minute!
Thankfully, not one corner teared from the stress of being folded together. That made me want to try that beautiful Pergamenta paper again. It had teared when during the tests on the vinyl cutter. However, since it was actually less sturdy than the Bristol paper, perhaps this perforation technique could be the solution for the Pergamenta as well?
I only had one small piece left that was somewhere between A4 and A3, which I taped to the laser bed. I did two tests to figure out the laser settings. However, while the laser was creating Resch’s pattern I could barely see the dashed lines. I was afraid that I’d set the laser too weak in the end. However, once it was finished and I started on the folds I noticed that the lines had been cut/etched enough to easily create the perfect fold lines.
Nothing teared during the folding and not even as I popped in Resch’s pattern and squished the whole pattern together to form the hexagons! (ﾉ◕ヮ◕)ﾉ*:・ﾟ✧
Using the “dashed line” technique really showed how well it worked; from flimsy to really sturdy paper, to paper that had teared in earlier experiments!
I then moved on to the engraving technique. I created a PNG of the SVG path in Illustrator and loaded it into LightBurn.
I quickly discovered that some of the straight lines in the pattern were a problem. When the laser came across a horizontal line it would cut along that entire section. To not cut through the paper I had to set the power quite low. However, at the non-horizontal lines that wasn’t nearly enough power to create “etchings” that you could fold along (I paused the laser to check after some time).
Also, engraving took forever! Michelle showed me a neat feature where you can see the estimated time of the job, which was 1,5 hours in my case (click the little monitor icon along the top I think):
Since the horizontal lines were an issue, I slightly rotated the image, and increased the power to 20% (speed of 200). It was again slow going, so after about 15 minutes I paused the laser again to check the results. Interestingly, the lines looked like the pattern was pixelated, or created with a matrix printer with a rather bad resolution.
Although all the lines were easily foldable, the line that had been the most horizontal (the bottom darker line in the image below) had been cut too far and teared while I tried to fold it.
I discarded this “engraving” route, mostly because it wasn’t scalable due to the time it takes, and these two tests hadn’t been very promising.
I also quickly did a test that Michelle, our lab assistant, had mentioned; to use two layers of paper, where the top layer would be “sacrificial” and stop some of the corner power.
I used a speed of 200 with a max power of 20%. However, once I took off the top paper (which almost fell apart) and looked at the one below I saw that the corners had gone through quite far already while the other lines weren’t even visible.
Perhaps I could’ve done another test where I’d increased the max power even further, but at the time my mind was done for the day after all these tests, and I’d had a good success with the dashed line technique, so I left it.
In general, all these tests are basically a result of the fact that the laser cutter’s power can’t be set to lower than 10%, which results in corners being cut through where the machine slows down. If I could put the (corner) power to 5% or even 1%, then I could’ve just loaded any origami pattern, fiddle with the exact settings a bit, and then let it run to create nice etchings while not burning through the paper.
Nevertheless, it turned this assignment into a very explorative and inventive collection of experiments to see if I could make something work.
Note | I do want to note a few things about the laser cutter in the Waag though:
- It’s a old model, about 10-15 years old (and that model was replaced with a new model 10 years ago).
- Henk modified the laser cutter completely! The most important update was that Henk replaced the controller. Now it is able to handle up-to-date file-formats (pfew!). Before, they had to use ancient file formats corresponding with the “state of the art” formats from 10-15 years ago.
- The mechanical and laser parts of the machine are still “old”. They’re doing fine, but might need replacement one of these days to bring it up to a decent standard.
What went wrong
Almost everything that I tried, from the laser cutter to the vinyl cutter resulted in paper that either couldn’t be folded, was teared too far already when it came out of the machine, or teared while making the folds.
What went well
Finally though, one of my tests succeeded! (๑•̀ㅂ•́)ง✧ Using dashed lines in the laser cutter resulted in paper that could be easily, and crisply, folded across the lines, while the paper didn’t tear; not in the corners and not while folding. The technique also worked for a wide range of paper types; from flimsy printer paper, to quite sturdy art paper, and even paper that had teared before during other tests (the Pergamenta).
What I would do differently
With the Pergamenta paper not tearing with the “dashed line” technique, it makes me wonder if perhaps the Cromatico paper could’ve also worked with the same technique. However, I didn’t have any more sheets to test on.
Although the “dashed-line” technique resulted in the best fold lines, you could still see the brown burn marks of the laser on the paper. With Resch’s pattern, which has mostly mountain folds, it was easier to keep the burn marks on the “outside” of the pattern, so you could see the fold lines better. However, if the folded paper should go on something that is being displayed, I should try to flip the paper around so the burn marks will be on the inside.
Perhaps I could’ve done one or two more tests on the “sacrificial layer” approach where you place a second layer of paper on top to stop some of the corner power. The test that I did hadn’t used enough power for the straight lines, which hadn’t come through the top layer, while already showing the corners. I should’ve done another test setting the max power higher.
Although I tried so many things this week, it was all to merely pre-make a fold pattern to do origami. I therefore have a weird two-fold feeling about it all. One the one hand I know that getting to that point took a lot of experimentation, and I’m proud of that whole route (and a big thanks to Henk for the many suggestions!). On the other hand, I feel like the end result isn’t all that special? I haven’t really made anything.
Well, I guess the sheets of tessellation-folded paper do look kind of pretty even though they have no use (✿◠‿◠)
All Files Together
- Optimized Resch’s origami pattern | SVG file