Let's Make Robots!

Christmas Star (3d printed parts)

http://www.flickr.com/photos/nand/sets/72157632264318254/

This year I designed, printed and assembled a Christmas Star for my parents' Christmas tree.

The star was based upon a similar one I had made from brass tube a few years previous and since my folks wanted it silver coloured I decided to go with aluminium rod for this one. I figured that since I couldn't bend and solder/braze the aluminium rod (as I had done with my brass star) I would 3d print the corner connecting pieces. I designed the corner pieces in OpenSCAD (two screen captures at the beginning of the video), an open source 3d modelling program. The model is parametric and I intend to share it with thingiverse once I have tidied up the code.

Printing of the parts caused me some headaches as I had trouble getting various 3d printers to work. In the end I had assistance getting the Ultimaker at MAKLab to work and the parts that it produced were of fantastic quality. Included in this video is a time lapse of the Ultimaker printing the remaining 8 parts (there are 12 printed parts which assembled with 20 alu rods makes one star).

I took some photos during the assembly stage and include in this video the failed attempt to use hot glue (too rigid, wouldn't allow me to pull the parts together to put in the final connectors). I briefly attempted to use zip/cable ties and fishing line but neither of these methods were particularly successful. Final assembly used wire wrapped around the alu rod and through holes drilled in the connectors, these joints were then superglued once the whole star was assembled.

Lessons learnt:

Sense of scale: When desiging parts for printing get a good idea what size they will end up. My small parts were too fine for one of the printers I had access to and a lot of the detail was lost. Also, lovely angled surfaces on the screen will come out stepped and depending on the layer thickness this may not be suitable.

Think through the assembly process: I hadn't given enough thought to how this was all going to be assembled. "I'll just glue the corner brackets to the rods" isn't enough, as I found out. I could have included some attempt at temporarily holding the rods in place while the glue set. I could have included notches or holes for the wire/fishing line/string etc but instead ended up drilling holes in the plastic parts.

Don't leave it so late: Timing on this project was pretty tight and I was fortunate that the printer behaved for me when it did as that was going to be my last attempt.

Designing for printing doesn't need to be ugly: In order to make my design printable I bulked up on the body of the couplings. Then when printing to the Ultimaker I found out that it does a pretty good job with support material. If I were to print these parts again I might go back to my original, lighter design and forgoe the heavyweight body. The printed parts are plenty strong, even with only 40% fill and if they weren't then some thin web material (like the web of an I-beam I mean, not a spider's web) could be implemented to strengthen in certain directions.