First Prints

 

My first print was actually pretty boring.  I made a box big enough to cover most of the platform and printed the first layer.

 

Covering the bed in straight lines is a good way to check the level.  If the the threads start gapping to one side, then it might be a little low in that area, of if it is too high they will squish together and cause little ridges.  This seemed pretty consistent, maybe a little light on the flow rate since you can see through to the kapton here and there.

The next print was Yoda http://www.thingiverse.com/thing:10752.  It’s more ambitious than some cubes, but I wanted to throw a challenge at it right away.  I ran it through Slic3r, which is usually my slicing software of choice.  It is much faster than Skeinforge, and took less than 3 minutes to slice, compared to 20 minutes in Skeinforge.

There was a lot of blobbing on the surface of this one, cause by the extruder pausing as it drew the perimeters.  Whenever it paused, plastic continued to ooze creating a little blob on the surface.  This might be due to the printer waiting for commands from the computer which couldn’t send them fast enough.   Slic3r seems to have cut the perimeter an excessive number of small segments, because the gcode was 11mb, as compared to the Skeinforge gcode which was 3mb.  This means Printrun needed to push a lot more commands in a small amount of time.  Later I will try it with another host like Repetier, and see if there is any difference.  Also Slic3r lets you set different speeds for perimeters and fills, so I could try having it draw the outlines much more slowly.

The next one was sliced with Cura, which is an all-in-one program.  It will set up the model on the platform, slice it, display the toolpath as lines in 3D, and then control the printer.  I still used Printrun for control, because I use an Atom based PC to run the printer which is far too slow for slicing.

Cura had less blobbing, but it made a mess of the fill.  This is the first time I have tried it, and it does some things differently.  Rather than tell it how many outlines you want, you tell it wall thickness and then it calculates the outlines.  Some of the solid fill layers weren’t complete, other times it looked like it drew over the same areas twice.  By the time I stopped it the solid layer was actually convex, which made the nozzle drag through the plastic.  You can see that happening in the inital layers as well, where the nozzle gouges tracks in the plastic as it moves between areas.

The last version was sliced with Skeinforge using the Solidoodle profile.

I only used two perimeters, which isn’t enough for this model.  There are some curves gradual enough that two threads don’t create enough overlap with the layer above so there are some gaps.  I also didn’t use support so there are some dropped loops.  Lower temperature might have helped that as well, ABS is less likely to droop if it is cooler.

For next post I will take a model through the software from start to finish, but I wanted to get a print out there first.