9. Textile as scaffold
Never thought I'd be making crystal mesh, Loes Bogers, 2019
To Do
- Document the concept, 3D model of the piece and document the design process
- Make two samples with 2 out of the 5 techniques to make a prototype of a textile form: fabric formwork with casting crystalization wood-textile resin-textile leather molding or other
- Document the process from CAD to CAMM, document how to use the CNC mill and prototype your textile composite
- Upload your 3D model and CAMM file
- Document the 2 processes you have followed step by step from design to machine and hands-on making, materials you used, your mistakes, failures and achievements
- Use 3D modeling software to simulate your fabric deformation
##Results: crystals, composites and (hopefully) leather forming
Conductive Alum Crystals on a pipe cleaner, tulle and conductive thread, made with Bare Conductive Ink, Loes Bogers & Frank Vloet, 2019
Alum Crystals on velveteen, felt and loosely woven cheesecloth, Loes Bogers, 2019
Single sheet silk chiffon bioresin composite, first try, Loes Bogers, 2019
##Growing crystals using yarn and textile
Sugar dissolves in hot water, but not in cold water. Dissolving: loose molecules spread evenly throughout because in hot water everything is dancing and agitated. When it cools down the molecules want to group together again, into a (partially) solid form. Crystals form better if they have a rough substrate to grow on, like a piece of thread or fabric. If you are using a dirty container or one with rough surfaces they can start growing on the jar.
The less you stir the water, the bigger the crystals can grow. Less agitation is better.
Alum crystal recipe
This is the source of the recipe we used.
- Alum
- Clean beaker or jar
- Saucer or shallow dish (a petri dish works, too)
- Pencil
- Fishing line
Process:
In one beaker, slowly add alum to 1/4 cup of very hot tap water, stirring to dissolve. Keep adding the alum until no more will dissolve: this is a saturated solution. Pour a little bit of this solution into a shallow dish or saucer and let it sit undisturbed overnight. Make sure you only pour the clear solution, not any of the undissolved material. You can pour it through a coffee filter if necessary.
Alum crystal growing on felt, after 2-3 hours, Loes Bogers, 2019
Seed crystals vs. growing on a substrate
The recipe talks about making loose crystals, but we're growing them on different substrates. I used velour/velveteen, a loosely woven cheesecloth, and a piece of tulle that I hung in small jars or put in a shallow dish (petri dish).
The next day you should see small crystals growing in the dish. When they look to be a good size, carefully pour off the solution.
To grow seed crystals Don't add a substrate, but instead just let the solution sit. Make another saturated alum solution with about 1/2 cup of hot water. Pour the solution into a clean beaker or jar; avoid pouring any undissolved material.
Remove the biggest and best-looking of the small crystals from the saucer to use as your seed crystal.
Tie the fishing line to the seed crystal. This can be tricky; a pair of tweezers will help. If you need to, you can score a groove in the crystal to hold the line in place.
Tie the other end of the fishing line to a pencil, then set the pencil across the top of the jar so the seed crystal is suspended in the alum solution without touching the sides or bottom of the jar.
Note: if your seed crystal starts to dissolve, that means your solution isn’t saturated enough. Remove the seed quickly and add more alum to the solution, filtering off any undissolved particles.
Cover the jar with a paper towel to keep out the dust and let your crystal grow until you are happy with its size. When you take it out of the solution, set it on some plastic wrap to dry.
If you see other crystals growing in the jar, transfer the solution and seed crystal to another clean jar.
The small crystals that formed in the saucer grew because of nucleation. A few alum molecules found each other in the solution and joined together in a crystal pattern. Other alum molecules continued to join them until enough molecules gathered to become a visible crystalline solid. (Chemists call that a crystal “falling out of” the solution.) If you left these crystals in the solution they would continue to grow, but they wouldn’t get very big because they would all be competing for the remaining alum molecules in the solution. Instead, you took one crystal and used it as the only nucleation site in the solution. It was the primary site for the alum molecules to join together, so the crystal could grow quite large.
Conductive crystals
Frank and I made a few samples to create conductive crystals based on this recipe from EJTECH he found. They made some beautiful examples that they then used as capacative touch sensors for different applications.
- 1x Bare Electric Paint 50ml jar
- Boiling water
- Alum (or the same amount of borax)
- Fishing line
- Wooden stick
- Substrates like textiles/yarns/pipecleaners
The Bare Conductive paint we used was a bit old but heating it up and diluting helped dissolve everything anyway. We used tulle, pipecleaner stick and conductive thread for these samples.
Frank pouring the alum solution through a funnel, Loes Bogers, 2019
Boil the water and add 100g of alum. Stir until super saturated and the powder is no longer dissolving. Add 1-2 teaspoons of Electric Paint (amount changes the resistance of the crystal!) and stir. Crystals form bigger when it cools down slower.
Wait 12 hours, take out the crystal, reheat the solution and add 3-4 more tablespoons of alum to saturate it more. Wait for it to cool down before putting the crystal back in or you might dissolve it. You basically give it more food to grow.
Wait another 12 hours and take it out. Continue the 12 hour cycle replenishing the solution with alum depending on how big you want the crystals to grow.
Threading tulle with conductive yarn
We laced the tulle with some conductive yarn to see if this makes it easier to solder the crystal/sensor to a circuit.
ELECTRIC PROPERTIES OF CRYSTALS
##Bio-composite and Leather molding
I really loved one of the wooden molds Anastasia showed in her presentation on Cuir Bouilli, and thought it would be feasible for me to create a similar design using Grasshopper. So I did! This is the inspiration piece :) As I was playing around with it I kind of liked a less tidy piece, one that maybe looks a bit more like a tumor growing than a collection of neatly arranged spheres.
The mold Anastasia showed in her presentation (maker unknown), Loes Bogers, 2019
###Design Process
Approach
- do a very simple design, but one I can build from scatch in Grasshopper
- design something with a 2-part mold (never done before)
- go through motions of machine again. I worked with CNC milling machine a few times before for Fabacademy's Computer Controlled Machining Week, the Molding and Casting week and for the Composites week. But it's been a while, and designing for this machine can be quite complex I think.
- Figure out the process of leather molding (never done before), maybe even do a biocomposite as a bonus using the same mold :)
First designs
- I tried remaking a design from lecture that I liked in Grasshopper. When you know what you want it's much easier!
- Thinking in molds: adding a box. Searching for Boolean Union and Boolean Split in grasshopper (spoiler: it's SolidUnion and SolidDifference). On hindsight I didn't have to make the boxes this thick, it would only be extra milling time, so I chopped them off later.
- Thinking in molds: offset for the negative mold part to allow for material thickness. I pressed the leather I want to mold between a caliper, it measured
Measuring the leather to determine offset of negative mold, Loes bogers, 2019
First grasshopper design, made totally from scratch! Whoa. Loes Bogers, 2019
Grasshopper issues to be solved
As I started to elaborate the design I was struggling with very long waiting times and beach balls...I started adding panels to each output to see if they were giving the output I expected and if it matched the input it was going into.
When I baked I got like 10 shapes instead of one. And the random number generator was creating 144 values where I was expecting 7. So instead of directly connecting the output of the ball locations I put a [List Length] component after it and used that as input for the random generator. This fixed everything! Just one bake, and no more waiting.
Then I assessed a design I liked carefully with [DimAlign] in Rhino. I checked that the deep cavities were not too steep, and most corners and negative space between balls can be done with 5mm milling bit. I know it won't be perfectly sharp between the balls for the positive mold but I'm hoping that the negative mold part can compensate a bit for it.
Unmillable parts of the design that needed to be changed, Loes Bogers, 2019.
I decided to populate manually using control points in Rhino, because had to adjust the positioning of the balls so the design could be milled with a 5mm milling bit. The spaces between the balls were often too small, and moving around was an easier way to control the position of the balls than the 2D populate component. The best one I could think of anyway.
I streamlined the "code" as much as I could, making things relative to one another where I could manage and where it made sense. I'm pretty happy with result, it works very fast now and I definitely feel a little more confident in Rhino/Grasshopper now. Small steps, easy does it.
Final design in Grasshopper, Loes Bogers, 2019
Final design in Grasshopper, Loes Bogers, 2019
Design of the molds, Loes Bogers, 2019
And this is the Grasshopper file here
#Milling on the Shopbot
Lucky me, I had used the shopbot before and could use my notes from back then as we are still using the same machine and software at the lab! I documented it very thoroughly for the fabacademy CNC machining week, the molding and casting week and the composites week where I also made a composite with textile. But gosh it's so long ago! It was really nice to get a refresher with Henk and the other fabricademers.
I forgot you need to click the milling bit into the collet for example. And how to go through all the software steps. But the fear for the machine was engrained enough to still remember everything! Haha. I'm not afraid of the machine, just a little nervous when using it, but healthy nervous that makes me very sharp.
Settings
I'd prepared my design with a 5mm flat milling bit in mind and could use the settings that the lab manager, Henk had saved in the software's library. We changed the settings a little bit to work with high density foam, by increasing the stepover for the roughing toolpath, and increasing the feedrate. Because the foam is so soft, it can be milled a bit faster, milling a larger surface per toolpath, and going a bit faster that when you're milling wood for example.
Thanks for the settings and the tutorial Henk, Loes Bogers, 2019
**
The finishing toolpath
For the finishing toolpath we agreed on a 50% step over at a pass depth of 2mm foam. The basic settings are 18K RPM spindle speed (to be set on the machine itself), and a feed rate of 120. The machine estimated a 2 hour job which seemed quite long, and then we figured the feed rate could go up a lot for foam.
Job setup for the roughing toolpath, Loes Bogers, 2019