Oh me, oh my...it's part three of Cloud Mine (which I've been secretly calling Pieces of Heaven).
In part one of this Instructable (if autocorrect doesn't stop trying to turn that into "intractable" I'm gonna flip) I described all of the little bits and pieces of information that I took in to inspire this garment, including music, poetry, art, technology, and museum exhibits. In part two I talked about all of the elements that go into making clothing including, but not limited to: sourcing of raw materials, labor, textile design, silhouette, pattern making, and of course I hinted at budgets (everyone's favorite).
In this installation I will talk about how to create a prototype or custom garment using a CNC knitting machine, which will do the lion's share of the physical labor. I call this a prototype because there are some extra steps that I would most likely take if I were to prepare this for mass production (grading, notching, flats, tech packs etc.). The garment I made for this project was more like a first draft or a first round prototype. It's pretty standard for me to make multiple versions of a garment as I get it ready for a client or for production. Three is definitely the magic number. Usually by the third try all the kinks are worked out (fit issues, design details, etc.). If I were to do it over again, I would definitely make a few changes. I will talk about those things as they come up, but on the whole I am very happy with the outcome.
Please enjoy the visual delight.
Ok, enough of that. I asked my knitter, Myrriah, to describe, in a few sentences, the process for making garments on her Stoll. This is the list she sent over:
Make a texture or color pattern
Create the structures and modules
Test the fabric and find the stitches and rows per inch
Create a shape and choose all the shaping modules
Position it on the fabric visually.
Process the pattern
Load it onto the machine
We've touched on a few of these, elements (making texture/color patterns, creating structures or pattern making, and test fabrics) in the previous Instructable (NOT INTRACTABLE), but I hope to get into these elements more deeply and describe the ways these processes interact with the knitting software and knitting machine.
But first...let's talk about knitting machines!
Step 1: Knitting Machines
If you haven't gathered already, knitting is old. Really old. Really, really old. I will spare you most of the historical details I've found, but the oldest knitted garment in existence (a sock...one sock...how do they always get separated?) dates all the way back to 256 AD. I don't even know when that was.
Today, home and professional knitters alike have a variety of tools to help them create knitted items in a fraction of the time compared to by hand. That sweater you bought from The Gap last holiday season, well, that was most likely made on a machine very similar to what I used for this project.
The video that's posted above accurately describes the technology I was working with and some of it's capabilities.
Step 2: So...Knitting Is a Digital Manufacturing Additive Process? Say What?
As it turns out I'm not the only human who's been looking at knitting machines and saying, "Hey, that looks a lot like a 3D printer!"
Knitting machines are, as described by the lovely lady in the above video, a digital manufacturing tool using additive processes.
For those of us who are new to this kind of terminology, as I was before I entered The Pier, additive manufacturing, "is defined as the process of joining materials to make objects from 3D model data, usually layer upon layer, as opposed to subtractive manufacturing methodologies."
This project falls short of this definition only because the data is not in 3D, at least not yet. It does have the layer upon layer thing going for it though. Generally speaking I make garments by draping a piece of fabric around a 3D form (in the real world) and translating that to 2D pattern pieces for construction and replication purposes.
I have to mention that traditional flat pattern making and draping processes can then be though of as "subtractive manufacturing methodologies," where the pattern maker takes a fixed width textile and cuts shapes from it that will eventually be reassembled to create a finished garment.
So now that we know what kind of process knitting machines use, let's look at a few brands of CNC knitting machines and the machine I worked with.
Step 3: There's a Machine for That...
Step 4: Make Texture or Pattern
Now we'll look more deeply into the process that turned my inspiration into form.
I gave a detailed look into the textile design processes for this garment in part 2. Even still, there are some additional points I can make here as we begin to talk about interacting with Stoll's software.
1. Because of the proprietary nature of the Stoll machines and their accompanying software, there is not an easy way to design textiles outside of their interface. This was a challenge for us. The closest solution we found was saving my design files as bitmaps (.bmp) and making slight adjustments once the image was processed by the software. Sometimes this meant altering every square inch of the textile, pixel by pixel, stitch by stitch, in order to achieve the desired effect. I have to give a big shout out to Myrriah for putting up with this at those moments.
2. Knowing what level of detail you want and will accept in your textiles is really important.
3. Adjusting the scale of the custom textile was relatively easy within the program.
Above you can see a screenshot of the sky textile within the Stoll interface. We were attempting to get the number of colors shown down to as few levels as possible, this is one of the process shots. If we had a larger gradient of yarn colors to use in the project we could have used this iteration.
Step 5: Create a Shape and Shaping Modules
The flat pattern that I created and measured in part 2 is translated into data points that the computer can understand and use while knitting the garment. I've included an example of what the data to communicate a pattern to the machine looks like.
Step 6: Position on Fabric
Once your textile is all set up and your pattern data is defined you can place the digital pattern (created with data points) on the digital textile in the design environment. I was able to move the top up and down along the length of the textile until I found a spot that "felt" right. We got some really great screen captures of the process. The section that was knitted is bright, while the unselected sections are in shadow. You can also see a photo of one of the test tops, complete with waste yarn.
Step 7: Process the Pattern
This part of computerized machine knitting is completely magical and entirely incomprehensible to mortals, as evidenced by Myrriah's inability to speak on this part of the manufacturing process.
When I asked if she could explain it to me the reply was, "Seriously, no one will understand this unless I get nitty gritty on the way the machine works. I can't do this part for you, sorry."
To me this means the machine is magic.
Total pattern programming time:
Approximately 6 hours (+/- 2 hours)
Step 8: Load the Processed Pattern Onto the Machine
This part of the making process is done at the machine itself, separate from the programming of the pattern and textile, which is done on a separate computer terminal. The saved, processed pattern gets loaded onto the machine either by LAN or by using a USB (thumb) drive.
At this point the machine operator can define elements stitch size, fabric tension, yarn feeders, yarn set up, floating in and out, knit speed, start, and finishing elements.
Above you can see the yarns all hooked up to the machine and going through the various feeders. The gold lurex has a slip to it, so the plastic bag is necessary to keep the yarn on the spool.
Step 9: Knit It
Here are a few short videos of what it looks like to run a Stoll knitting machine. It's a copier for clothes, a printer for pants, an undeniable digital manufacturing additive process machine of wonder.
Skirt - 2 hours 15 minutes
Bodice - 6 minutes
Pockets -10 minutes
Ribbon Trim - 1 hour
Total Knitting Time: 3 hours 31 minutes
Step 10: Test the Fabric/Find Guage
I put this process at the end because for me this step can act as a reset for the whole project, depending on the results. You pretty much have to go through all of these steps just to knit swatches. After washing and drying the fabrics and determining if the density and gauge (stitches per inch of fabric) is correct for the garment, the technician can make adjustments and either knit the whole garment (if everything looks good), or knit more swatches (if things need to be tweaked).
Step 11: Linking It All Up
We are so close to a finished garment it's not even funny. But you know what is funny? The sweet muzak providing accompaniment to this promo video for a professional grade linker.
With knit garments the final seams get linked together instead of sewn. This insures that they maintain some amount of stretch and look consistent with the rest of the piece. The outcome is also invisible when done correctly.
I've also included a video of Myrriah linking a sample from our project.
Total finishing time (linking and weaving in ends):
Step 12: Blocking
Most knit garments need to be washed or "blocked" before they are worn or sold. Blocking gets all of the stitches neatly snuggled up against each other, relieves any irregularities in the textile, and allows the maker to give the garment a final shaping.
Step 13: Futures in Digital Garment Making
Now that we've gone and made a fully functional garment using a computer controlled knitting machine, I want to talk about futures I can envision in knitting technology and how that might change our actions as both consumers and makers.
One comment that I heard a lot around the pier while people interacted with the various machines was how much more convenient it would be if the programs that ran the machines were available for use on our individual computers. Like Myrriah's knitting machine, there are some processes that can only be defined while the user is working with the proprietary software . For example, there was some tedious design work that I completed on my own that Myrriah had to replicate because her software couldn't interpret the data accurately enough. Making certain features of the software available to general consumers, so they could do some of the heavy lifting on the design side might change the nature of garment making considerably.
Just imagine if I as a consumer or maker didn't have to own a knitting machine to design using it's interface. I could learn how to program the machine, make textiles and patterns, save them in the proper format, and then send the job to someone who has the machine. They could "print" out my garment, link it if necessary, and send it to wherever it needs to go. As it stands there is at least one other group of thinkers that is figuring out how to connect people to custom made textiles.
Myrriah's production time for the CNC knit garment is comparable to my production time using analog tools - pencils, pattern paper, rulers, pre-made textiles and sewing machines - around 15 hours. This means that in this case there was no significant reduction in overall prototyping time while using the technology. The benefit of course was that I could design something to my exact specifications. I created the textile itself, which is no small thing.
Customization through technological interaction seems like the next wave in fashion, at least it is here in the Bay Area. As long as machines continue to run with the help of computers, this will be the case. As I started this residency I connected with some entrepreneurs who wanted me to help them get their customizable leather jacket company going, and I learned about this shoe company that also utilizes creative web interfaces and customization as bait. I think the success of these companies is in the UI, because customization in garment making isn't a new concept.
Just because our requests are going through a website doesn't mean that this is the first time in history that people have gravitated toward fully customized clothing. This work has been done for eons by women and wives, seamstresses, tailors, and cobblers. Most of the work is still done by skilled professionals that have spent considerable time learning the trade and industries, the only difference between then and now, like so much in America, is the packaging. New packaging = profits. The methods of production haven't change significantly yet.
The pictures and videos are meant to express some of my concerns with the upcoming custom fashion craze. "I dress myself!" Us designers make sure you don't end up like poor Ralph, or wearing a shirt that closely resembles innovations made to cover bald spots (although there is something fun about anything spray on).
Step 14: Remember, a Goddess' Robe Is Seamless
A goddess's robe is seamless for it is woven without the use of needle and thread, entirely on the loom. The phrase "a goddess' robe is seamless" passed into an idiom to express perfect workmanship.
- China, Tang Dynasty
In my studies I've found that almost every ancient culture honored weaving, birth, death, joy, water - the very essence of creation - though the goddess. A god of weaving doesn't exist in mythology, and in the rare case that it does exist it's usually considered a theft, a misrepresentation of the archaic role. I am not trying to say that it is a woman's duty to be in the sewing room making clothes for everyone (goddess forbid), I am merely pointing out that ancient symbolism consistently identifies the weaver goddess as a container for existence itself. She is the gateway to all consciousness.
My concern with future technologies is that in their attempt to "eliminate inefficiencies" they may negate a rich and complex history of fabric and garment making that forms the foundation of all civilization, and along with it the livelihood and culture of women. Our histories are recorded on warp and weft, encoded in the amalgamation of knit and purl. If the mythology holds true, such an exclusion is hardly possible.
I've been pondering the fashion industry for some time now...what can I make of the scantily clad models that grace glossy magazine covers and billboards? Do they help or hurt women's sense of identity and power? Isn't sexuality a power? Doesn't mechanization and automatization merely a reflection of the reproductive potential inside every woman? How can a woman embody femininity and command the respect that she deserves? Is sexualized imagery actually degrading?
I don't have any concrete answers to these questions, only a shift in perspective. Perhaps fashion is the place where the goddess has been able to survive, a corner that she's willingly inhabited to avoid extinction. Maybe she's been playing dumb this whole time, waiting for the right moment to come out of hiding, waiting for someone to crack the surface and let her out again. Who knows? Only time will tell.
I hope you've enjoyed!