1. The 45 degree rule - "[FFF] printers do not print with dissolvable support material, so whenever possible," you want to avoid support material. (Overhangs that are greater than 45 degrees will need support material.) So keep your layers within that 45 degree or cut up the design and print the cut parts as self-supporting, then assemble. (see Plated Okapi below)
2. Add custom support material - don't rely on the computer algorithm - "be smarter than the computer" and design your own.
3. Ditch the raft - Rafts are "good for beginners" (like us!) since it helps with adhesion and helps level out the build surface if the build plate is not quite level. But it adds print time. If you have ditched the computer-generated support material, why not ditch the raft? Add "mouse ears" or helper disks instead (shout out to TinkerCAD)
4. Know your details - "double your thread width + fudge factor" = what is the smallest feature you can design? You need to know the limits of your machine.
5. Design in fit tolerance - She says this is tough. She recommends making a test print with many different diameters to see what fits in, so that you can design in the correct measurements.
6. Use shells properly - shells are "copies of the perimeters" or the walls. Two shells is good for most prints, unless you think your design will need to withstand a lot of stress. Be careful not to use too many shells in a delicate areas, as this may prevent any kind of fill.
7. Optimize walls for thread width - Take advantage of your thread widths to optimize print time for thin walls.
8. Orient for best resolution - Resolution only refers to z-height. Thread width (generally) does not change - either .4 or .5 mm (measurement of the nozzle). "It is always going to be the same width, no matter how thin you can make it." So orientation matters!
9. Orient for stress - Prints can break along layer layer lines, particularly with ABS. So again, orientation matters!
10. Tackle print in place - "pull the element designs to the build platform," build out at 45 degrees, "use bridges for captive parts" and "gap print carefully" (<< this #10 tip I'll need to investigate more, since I don't fully understand what she is talking about) - Check out the thought process in orienting this print, Umbrella.
Earlier in the talk, Hultgren interestingly comments on iteration saying that tries to "prevent iterations" (9:53) "We always say it is good to iterate on a 3D printer - it's even better if you can get it right the second time, or maybe the third time, instead of the tenth time." In the Q & A, she gives this advice: "I used to sort of create a design and then test it until it worked. And now, I keep all of this in mind before I start and I do as much testing upfront as I can, so that once I send it to printer I am pretty confident that it's going to go ahead and print."