Filament Extrusion: The Art of Trial and Error

March 30, 2014 3D Printing

In my last post, I showed how manipulate an STL mesh to make it more blended and organic.  I ended with pictures of a scale model of a test tube vase.  Looking at the pictures below, you can clearly see that my I need to play with the settings of my extruder.  While it printed, it is not pretty.  There are blobs on the edges, as well as a lot of strings that I had to clean out after the print.

  • Finished print… after cleaning out the cobwebs.

So, I set out to fix this, and figure out better extruder settings for my Makerfarm Prusa i3 with a 0.35mm J-Head hot end.  First, I created a small model that I could print quickly to compare the effects of tweaking different parameters.  I wanted to make it complex enough to emulate things I’ll be printing in the future, but simple enough to limit it to a print time of <30min.  Here’s what I came up with:

  • Creating the framework based on incrementally rotated hexagons.

I wanted this test piece to be printable with one outside layer, so I made the thickness 0.4mm.  I sliced it and then proceeded with my experiments:

  1. Original settings.  Extrusion temp 205 degrees C, bed temperature 70 degrees C. Retraction 1mm. Layer height 0.25mm.  Result is blobby… like my print above.
  2. Temperature test.  Extrusion temp 195 degrees C, bed temperature 70 degrees C. Retraction 1mm. Layer height 0.25mm.  The print is a bit better, less blobs and strings.
  3. Temperature test.  Extrusion temp 185 degrees C, bed temperature 70 degrees C. Retraction 1mm. Layer height 0.25mm.  This was too cold, as indicated by layers that were not fused together.At this point I determined that 195 was the proper extrusion temperature.  Doing some reading around the internet, I found that a lot of people use this as their extrusion temperature.
  4. Sample 4: Retraction test.  Extrusion temp 195 degrees C, bed temperature 70 degrees C. Retraction 2mm. Layer height 0.25mm.
  5. Sample 5: Retraction test.  Extrusion temp 195 degrees C, bed temperature 70 degrees C. Retraction 0.5mm. Layer height 0.25mm.
  6. Sample 6: Retraction test.  Extrusion temp 195 degrees C, bed temperature 70 degrees C. Retraction 0 mm. Layer height 0.25mm.While 0.5 mm seemed to have the best results, it was still not great.  I looked at some other blogs and found that they were using a layer height of 0.2mm instead of the 0.25mm that was default for my printer.
  7. Sample 7: Layer height.  Extrusion temp 195 degrees C, bed temperature 70 degrees C. Retraction 0.5mm. Layer height 0.2mm.  Eureka!  This print was much better.  The surface finish was nice in most areas, however there were still some thick areas around the hole cutouts.At this point, I looked a bit closer at my model, and then at the output from Gcode.ws.  I realized that the blobs were actually an artifact of the slicing.  Looking at the layers, you can see little “Y” paths at the edges… so this is what was adding to the blobs all along.
    GcodeWS
  8. Sample 8: Layer height.  Extrusion temp 195 degrees C, bed temperature 70 degrees C. Retraction 0.5mm. Layer height 0.2mm.  For this one, I suppressed the hole feature and stripped it down to just the surface and used the “Spiral Vase” option.  The print came out beautiful!
  9. Sample 9: Layer height.  Extrusion temp 195 degrees C, bed temperature 70 degrees C. Retraction 0.5mm. Layer height 0.2mm.  Here, I modified the wall thickness of the model to be 0.8mm, so that 2 outer layers would be formed.  This model also came out very nicely.  Minimal blobbing, and spider webs.  Looks like I found my new settings!

  • The full lineup, in order from 1 to 9, left to right.

Here is a close-up of sample 8:

A_DSC_0159

So, while not a true Design of Experiments exercise, I accomplished my goal of improving print quality.  From now on, I’ll be printing my PLA model at 195 degrees C, with a bed temperature of 70 degC, retraction length of 0.5mm and layer height of 0.2mm.  I’m excited to try out some new prints!

Leave a Reply

Your email address will not be published. Required fields are marked *