Individual assignemnt
cut something on the vinylcutter
design, lasercut, and document a parametric press-fit construction kit, accounting for the lasercutter kerf, which can be assembled in multiple ways
Group assignment
characterize your lasercutter, making test part(s) that vary cutting settings and dimensions
To get a fancy sticker out of the plotter cutter I had to prepare an image file that would be prompted to the machine along with vynil adhesive tape. This process is composed of different steps, depending on your need for the final result, but generally it’s a good practice to start from a SVG editor like Inkscape to work on the desired file. I’ve already exposed several notions about working on Inkscape on a previous assignment, and in this one I generally use the same tools and techniques to obtain an high resolution PNG file. To start, I downloaded an SVG file from a web gallery called The Noun Project: the lovely silhouette of a LEGO man. SVG file
I resized and aligned the image to the page, then I proceeded to export it as a PNG file via File -> Export PNG Image...
menu command. The tab that opens up with this instructions prompt you to fill some technical detail, such as the dimension of the export area and the DPI desired. In my case, I made an export of only the Lego man and a 500 DPI resolution (this high amount will grant that the curve lines of the cut will be very smooth an natural).
With this new PNG image I proceeded to upload it in the CutStudio software, which is the proprietary application to run the Roland plotter. Once uploaded, the software behaved strangely and wasn’t loading properly the image file, since in the editor window it showed only a grey rectangle. The only message given by the program was a suggestion to darken more the color of the image. After some scanning of the Inkscape settings and a lot of headscratch I was able to find the problem in the color standard used to rendere the PNG file. CutStudio required the file to be formatted in RGB while my Inkscape environment was set to HSL.
After switching the palette in the Fill and outline
tab and exporting again the file, the image finally got accepted by CutStudio.
To analyze more results with the same image, I doubled the image on the editor window of CutStudio and surrounded one with the Rectangle
tool. This technique will be used to create a negative of the same image by peeling off different parts from the not-contoured image.
After adjusting position and size of the images, it’s important to use the command Image outline
appearing with the right-click over the figures you want to cut. This is required to define the proper cutting route for the plotter, otherwise the machine won’t do nothing .
I switched on the plotter cutter and fed it with a square of vynil adesive tape big enough to hold both images.
Once the machine was set, through CutStudio I proceeded with the Cutting
button which opens up a settings window. In here it is possible to change many variables such as the strength of the blade and the speed of the head, but since the machine was already properly set I only pressed the button Get from machine
which not only confirm the values set in the cutter, but also recognize the size of the tape inserted into it via photosensor and a couple of rolls you can adjust to define the precise portion of area to cut.
If the settings are set, it’s possible to go on with the cut through the Ok
button and the machine will start automatically to cut.
A few minutes of scratches and buzzes later, with the adhesive back in my hands, I had to peel off only the area I didn’t want to result in my sticker, which is a process that requires very fine motor skills (possibly with the help of pliers or a scalpel) or you risk to ruin the whole design. Once the unwanter parts are removed, I carefully attached a transparent film on the vynil tape.
This film is used as a transfer to get the sticker off of its paper slot and eventually attach it wherever needed.
The laser-cutter machine we have in our lab is a Trotec Speedy400 Flex which is a very versatile tool. All of its specifics are enlisted in its official manual, which was indeed a useful reading to understand what it does and how it does it.
Basically its a CNC laser cutter with two laser (CO2 and fiber) with a maximum power consumption of 120w. I can cut over surfaces up to 1000*610mm (with a surface height ranging from 125 to 305 mm) but frontal pane can be lowered and bypassed to allow longer piece to fit and be cut.
Since for this assignment, I was given a brand new plywood board, I followed the same process described in the group assignment to calculate the kerf of the laser cutter on this undocumented material. This time, however, I focused on a more precise range of power and speed ration, since the previous experiment already provided some indications of how laser behaves with wood. To make the test, I prepared a grid of 5 different settings and I proceeded to cut them.
After the cut, I inserted the file inside a spreadsheet to calculate the best kerf with the best speed/power ratio and by far the best result was a 0.17 mm kerf obtained with a setting of 80% of power and 0,4% of speed.
Given some more time it would have been interesting to test more in the range around this result, since it happened to be the last element of my matrix. It might be a good test for some future experiment.
Since my project for this assignment will be composed of various elements that in my mind will intersect only transversely (meaning that every joint will be put on the width border of the other elements) to calculate the best length for the pressfit to just stay in place I wanted to craft a tool that would allow me rapidly to find the best measure. That’s why I proceeded to draw on Inkscape a circle with many rectangle punched in its whole circumference, each one with a slightly different length. This process has been quite straightforward and repetitive, since I just had to clone a rectangle 8 times, rotate it of 45° for every step and justaposing it to the circle.
An important thing to know is that when you work with clones on Inkscape it’s very important to unlink them, otherwise any change to anyone of them will be executed accordingly by all the other clones. To do so, you have to select your clones and then execute the menu command Edit -> Clone -> Unlink clones
.
After resizing the rectangles, adding them to the circle is the matter of a boolean subtractive operation, which allows to carve away shapes out of other shapes. This can be done, after selecting the desired shapes, with the menu command Track -> Difference
.
Keep in my that in the Inkscape environment if you made shapes with tools instead of drawing them out of lines, you have to convert them into tracks otherwise the boolean operations won’t work. This can be done with the menu command Track -> From object to track
,
As I wrote, this process was a bit cumbersome, and I’m quite sure that there are easier ways to accomplish this task, but they require more knowledge on Inkscape from my side.
With the SVG file ready I proceeded to cut my gauge with the laser cutter and test it.
The first result actually came out badly and wasn’t very functional, since I erroneously considered the thickness of a different plywood board from the one I would have used to make my project and I didn’t consider properly the kerf. It turns out that a little less of a millimeter is enough to prevent two pieces from fitting. After refining the file, I made a new cut and this time the result was very satisfying.
After few measurements, I found out that even a 6,10 mm hole can fit well a 5,60 mm thick plywoon and hold it quite well, however to not lose grip after some wear and to justify also the use of chamfer I opted to adopt a 5,90 mm wide joint hole (also, the plywood is quite irregular and has a few variations of thickness in some parts and this is highest value I measured all along the board.).
My idea for this assignment was to craft a kit of pieces that you can assemble in order to create funny, sketchy and modular animals. For this reason, I proceeded to create a various parts that could be intended as limbs, paws, torsos, heads and so on that all could fit together no matter the way you use them. As on hommage to Oulipo, I’ve chosen to name this kit BêPots (as in Bêtes Potentielle).
To simplify this idea, I’ve chosen to join the pieces only woth transversal axis, meaning that the two board thicknesses would always fit in a cross joint. This surely reduced the amount of time spent in the design process and cutting attentions, since I only needed to draw female parts in every piece.
To draw a 3d model of this kit, I’ve chosen to use clodu-based CAD software Onshape, which main features and functionalities were described in my previous assignment.
Since I had to draw a parametric model, the first thing I started to plan out were the variables that would later define the dimensions in the sketch. To do so, you have to press the Variable
button in the upper bar and prompt a name and a value.
Keep in mind that this values can be changed in any moment, so to make even different kits from different materials and different sizes.
Once all the variable were declared, I started to sketch out the 2D silhouette of each piece with various tools to make many kind of shapes. For each piece, I would define all of its measures with the Dimension
tool of the upper bar. This tool makes you choose a dimensione, such as the side of a square, the lenght of a segment or the radius of a circle, and then define it. This definition can be literal, with a pre-set number, or dynamical, using a variable or a formula.
An important thing to know is that in the Feature list
of Onshape the variables must be hierarchically put before the dimensions you want to use them in, otherwise they cannot be seen by the environment.
After defining drawing all the pieces, I also made a inlet object set out with its relative variables. This would be used later to subtract its shape from all the pieces so to make joint inlets inside them. For this purpose, I resorted to the technique of the chamfer, meaning creating a sort of triangle that eases the insert of a joint inside a material, especially rigid ones like wood, and helps it to stay in place.
Note that I used the Mirror
tool to replicate a side of the shape and quicken this task.
After all the shape were sketch out I proceeded to extrude them using a variable for the material thickness as width for the operation.
After this operation I made many copies of the inlet part and put them in the points over the other pieces were I wanted to make the pressfit slot, then I used the Boolean -> Subtract
command to cut away a negative of the inlet (exactlu as done in the thickness gauge test).
Furthermore, here and there I added some more chamfer just for aesthetical reasons, and the final result is an actual kit of all the pieces.
Since Onshape doesn’t allow to export the vectorial file of many separate parts together, and sice I wanted not to download a file for every piece and the glue them all together, I came up with a gimmick a made the sketch of a board larger than all the set of pieces. After extruding it, I made another boolean difference with all the parts and the result is a board with every piece cut out on it.
Then, I right-clicked the surface I needed and selected the command Export...
. This process gave me the dxf file to download.
Handling a dxf file was more complicated than expected. Importing it in Inkscape resulted in a strange file with odd, tiny measurements and broken lines.
Not wanting to re-touch it, I tried to open it on Adobe Illustrator and it was still rendered as with tiny proportions, but this time of a 1⁄10 factor that much more reassuring to work on. After resizing and properly coloring the layout in red (the reference color for cutting in our lab) I proceeded to calibrate the laser cutter and initiate the cutting.
Aside from the excitement, the operation was a bit of a failure. In fact, the inlet joints were not large as intended and no piece could fit with another. After some thorough inspection of the Onshape model I realized that I made a huge and naive mistake: in declaring the kerf variable I confudes mm with cm and in the end the cut was much, much more narrow than intended. After fixing the issue, re-exporting the dxf file, re-touching it on Illustrator and sending it to the cutter, I finally was able to hold in my hand a properly sized kit of pieces that could fit together just fine.
This is the sticker resulting from the the vynil cutter work attached to back of my sketchbook.
One of the men is frowning because the mouth strip fell off of its position and I had to relocate it manually. I felt it was good this way.
The following are a couple of BêPots examples crafted by lab peers.
Unfortunately one of the paws broke while assembling it, this means that perhaps dimensions of the pieces must be a little bit larger to keep them less fragile.
I will totally full my surrounding with customized stickers in the next future. Beyond that, since the vynil cutter is a rather simple and straight-forward tool to use, I think I will want to explore imaginative and experimental way to use it and create graphics from scratch.
Regarding the laser cutter, it will surely be a fundamental part for my final project but since I think that using it properly requires a lot of experimenting, so I think I’ll want to try it with many more models and pressfit kits. Especially, I want to delve more into living hinges technique and try to cut many kind of material. The BêPots kit need a lot of refining and resizing (certain pieces looks really fragile and are too small to be handled easily) and adding some more pieces and ornaments could be the perfect gimmick to learn more on how to use Onshape and perfect the many bottleneck I’ve met in this work. Probably, BêPots of many shapes, size and strucutre will infest our lab soon.
On the following weeks, more BePots spontaneously appeared in our lab!