// Cambridge, MA

week 4: 3D printing

as i talked about in my final project planning, my goal this week was to take some initial stabs at 3D printing fabric.

following a few guides online, primarily this one. there are several ways to achieve a fabric-like textile via 3D printing. one is by making 'chainmail' via small interlocking pieces, but i found the results of this to be very obviously 3d-printed. two other methods involve editing gcode directly or making custom textile textures in CAD, both of which seemed a bit advanced for a starting point. so, i elected to go with the infill method, which i'll detail below.

i started off with a simple rectangle, 50 mm by 50 mm with 0.3 mm depth. my test prints would all take this basic shape; in textile world, this small fabric sample is known as a 'swatch.'

then, the key part of the infill method involves taking the thin rectangle, and removing the top and bottom layers in a slicer. this leaves only the infill, and i experimented with different patterns / percentages filled to see what worked best.

i first tried a gyroid-shaped infill, seen below with its resulting print:

after a few more tries, i found the most effective pattern to be quarter cubic, since it kept its shape in two dimensions and offered solid structural support while remaining decently flexible.

next, i wanted to see how a design would look when added to the swatches. i thought that 3D printing provided a good chance to try embossing, or the technique of having a design be the same color as the background material, but raised to provide relief. this has been made popular in recent years by brands like fear of god:

i chose to call my budding 3D-printed brand 'Christian 3Dior.'

the print came out pretty satisfyingly.

there's still work to do, but i was relatively happy with the swatches and decided to scale up a little bit to print the actual pattern of a t-shirt. i made an extremely basic pattern for a tank top, and printed using the infill method:

but the print failed :( -- at the scale larger than a swatch, it seems like i'm going to run into layer adhesion issues and warping. tbd if this fixable but i am continuing to explore.

i was running out of time for the week and also needed to accomplish a 3D scan. i chose to scan my roommate's shoe because it was the first thing i saw when leaving my dorm tuesday morning and i realized i needed to grab something to scan.

i found the leo scanner to be powerful but a bit finicky to use. i tried doing multiple scans to get the top and bottom of the shoe at different angles, but this didn't work well, so i decided to just suspend it in midair and try to capture it in one 360-degree scan.

i captured it pretty well, but definitely not perfectly. if i'd had more time this week, i definitely would have spent more time with this device to master its nuances.

overall, not a completely succesful week, but not a total failure. i made some good progress on proving that it's possible to print something that plausibly resembles fabric, and embossed designs work pretty well on it. but there's still more work to do, i.e experimenting with gcode or modeling textiles in CAD from scratch, which may do better printing at a larger scale.