Sunday, June 28, 2015

3D Printing for my Birthday

Every year on my birthday I do something I've never done before. This year I created a 3D-printed sculpture. I created it at the Proto Build Bar in Dayton, OH. The experience was a gift from my wife.

Rather than print a pre-designed piece, I made my own design.

"Spiracle Notice" - Darren Kall 2015
I made this Line Pattern sculpture. It is about two and a half inches high, five inches wide, and four inches deep at its extremes. I named it "Spiracle Notice".

I started in my studio making a CAD drawing of the sculpture in Tinkercad. I knew I would do a bunch of exploration with the drawing and didn't want to take the time to do the drawing at Proto Build Bar but they have the capability to do similar drawings onsite.

I used Tinkercad instead of the other CAD programs I've used because Tinkercad was recommended specifically for 3D printing. Tinkercad was very easy to use. It lacks the sophisticated features of more elaborate CAD programs, but it was perfect for what I had planned. 

My goal was to transpose one of my Line Pattern painting series ideas into a 3D sculpture. 

In my first attempt (#001) I translated my painted lines literally except I gave the lines depth. Drawing #001 didn't work well enough to even keep the drawing! My lines were laid down in a rectangular "frame" area much like I do with my paintings. The lines extended to the edges where I "cut" them to fit in the frame. In the 3D printing I would be limited to using only one color so I made sure my design would work in one color. Because of this restriction, the lines didn't really distinguish themselves enough to make the 3D printing interesting as a pattern. In my disappointment I also realized I was not taking full advantage of the sculptural space since all the lines were on the same plane. I deleted drawing #001.


Line Pattern drawing #002 - Spiracle Notice

My second attempt (#002) used lines of all the same length, height and width. I manipulated the overlap and intersection of the lines in 3 dimensions instead of the single plane of the Line Pattern paintings. In Line Pattern paintings I create the illusion of depth by overlapping the lines. In the sculpture I had actual depth to play with.







Line Pattern drawing #002 - Spiracle Notice



This is a view of #002, the same CAD drawing, as seen from above where the overlaps and displacement are more obvious. The Line Pattern paintings are explorations of how lines intersect. The Line Pattern sculptures explore the same but add new challenges.

I spent a lot of time ensuring that all the lines connected to at least one other line. With the CAD drawing it was easy to spin the drawing in every direction and to zoom in and out to inspect each intersection. All the pieces had to be attached to make it a single sculpture and to prevent "floaters" or separate pieces that would just fall out when the 3D printer made them. Things I don't have to think about when I paint Line Pattern paintings.



Line Pattern drawing #002 - Spiracle Notice
Here is another view of #002. It shows how a simple rotation of the drawing creates a new pattern. 

I wanted to create a complex pattern, one that would test the capabilities of the 3D printer, be an interesting extension of my Line Pattern paintings, and be an aesthetically pleasing grouping.







I made several additional drawings that were variations of #002. I used different numbers of lines, different arrangements, and different patterns of connection. I even started one where I was varying the width of the lines. I was starting to play and create too much! This was fun. But to ensure I didn't spend my day making drawings, I forced myself to pick one for the 3D printing experiment. I liked #002 the best. 

I exported my drawing from TinkerCAD in four formats since didn't know what format was needed by the 3D printers at Proto Build Bar.

At the Proto Build Bar I used a MakerBot Replicator. I imported the right format of my CAD drawing into the print-time estimation software. The manager at Proto Build Bar gave me a lesson on the software. My task was to optimize the orientation of the drawing to minimize the 3D printing time. The sculpture could be printed in any direction but some orientations would take the printer longer to print than others for the same sculpture. And since time is money on the MakerBot I wanted to print in the shortest possible time.

The print-time estimation software has algorithms that determine the need for supports and printhead travel time. Optimizing is a manual process where I repeatedly changed the orientation and then ran an estimation. I started thinking through how this process could be automated to make it faster and more accurate than my trial and error process. Then I remembered it was my BIRTHDAY and I shouldn't be working :^). So I took off my software designers hat and got back to more trials, and more errors. I got it down to the best duration I could.  Then I transformed my Tinkercad output into the MakerBot proprietary machine-readable format. The Proto Build Bar manager started the 3D printer. Then we went away for a few hours.

Actually my wife, our son, and I didn't go far. We moved over to a soldering bench at the Proto Build Bar and as a team we soldered and assembled a build-kit for a handheld Simon-like electronic game. It was a lot of fun but we kept peeking at the sculpture during the process.

This is the final product as it came out of the 3D printer with its supports attached. Supports are necessary for pieces of the sculpture that extend too far out. Because of the way the printer lays down the plastic if there were no supports the plastic would not stay in the desired spot in space or would simply fall down.
The biggest pieces of support were easily pulled off since they are only lightly attached to the sculpture. Some others needed to be cut off with an X-Acto knife.
Even after the majority of supports were pulled or cut off there were still thin spiderweb-like filaments left over from when the printhead moves to a new location. I used tweezers and a magnifying glass to pluck these off.  

WHAT I LEARNED: It was a great experience. I can see 3D printing will be useful as a maquette maker for some of my sculptures. The surfaces and quality are even good enough to be a finished product maker.  My shapes were deceptively difficult for the printer modeling algorithms. The finished product looks simple but with all the intersections and "hidden" spaces the supports didn't always work. There were two areas on the underside where there were ends of lines that sagged. They probably should have had supports. Good feedback for me when I design the next one until the algorithms catch up.