I got training on the 3D printer this week. It's nice to know that EDS' Stratasys has dissolvable supports, though I didn't get to use that. It seems that it could be useful.
I spent a bit too much time thinking about what to print, but I then settled on what my mind randomly: a hollow-capped mushroom. Why that? that was something I couldn't make subtractively (especially since there's a bottom end of the cap with holes). After more difficulties downloading FreeCAD, I borrowed a friend's computer that already had SolidWorks on it and printed the little mushroom on one of the Prusas.
I am pretty bad at 3D modelling, but luckily the mushroom shape was somewhat easy to make. It didn't help to not be experienced in using SolidWorks, but I got what I needed to do done, despite Anthony curious on why my mushroom was not radially symmetric. Though I printed the mushroom with minimal supports settings, there still ended up being a lot of support material to remove. Removing supports with hands and tweezers is an oddly therapeutic way to quiet the mind and busy the hands, perhaps just as much as soldering is.
I thought the 3D scanning system was very cool. After trying to scan a pile of pens (don't know why, but I thought it would be interesting) and then realizing the pens were probably too shiny to scan, I decided to try to scan a paper crane. I folded a paper crane on some graph paper (for some reason, my head thought the graph paper could make things easier to scan, but in hindsight it might have been a bad idea) and put the crane on the rotating platform.
Despite thinking a well-defined paper crane would be easy to scan, it didn't work out. I got frustrated, trying for at least an hour to scan this thing, but it kept freezing and then making multiple scan copies resulting in a warped figure. It always happened when rounding the wing, so I wonder if there's some issue with the depth of a feature and limits to that, because the wing jutted out not insignificantly.
In the end, I ended up giving up on the crane and scanning the sample object, a miniature bust of Julius Caesar. It worked just as it worked in the training demo--that is, smoothly and pretty darn well.