fatboy1271 wrote: ↑Fri Jul 31, 2020 6:26 pm
If I print just a single cube it is much less pronounced, but it still doesn't measure evenly.
Some additional "material", maybe it can help somewhat. Remember, using the slicer you can control almost every aspect and
decide what should happen.
When looking for problems with retracts I'd use something like this:
3 simple boxes (20mm x 10mm x 20mm). Each one rotated at 22.5° to make sure the travel paths later on will go across
perimeters not only at the points where retracts happen (start and end of a layer/island):
www.aunix.de/tmp/s3d/s3d_retracts_1.jpg
I chose 3 perimeters and no infill / top layers. Don't really need them to check retracts, we can save some material and time here:
www.aunix.de/tmp/s3d/s3d_retracts_2.jpg
After setting up a single process I created three additional ones. The only difference between them:
1. Each process starts and ends at another height (0-5mm, 5-10mm, 10-15mm, 15-20mm).
2. Retract Distance (0.3mm, 0.6mm, 0.9mm, 1.2mm)
www.aunix.de/tmp/s3d/s3d_retracts_3.jpg
No coasting and/or wiping, no extra restart distance. Layer start points are aligned to the side of the bed where I
put the camera.
This is what it looks like with all the travel paths shown.
www.aunix.de/tmp/s3d/s3d_retracts_4.jpg
Now select "Preview by Line" inside the preview and check how the toolhead moves:
www.aunix.de/tmp/s3d/S3D_Retracts_Preview_1.mp4
As you can see it will do the same moves all over again. Start at the inner perimeter, print 3 perimeters, retract and move on to the next one.
This is the center one (already in the 0.9mm process):
www.aunix.de/tmp/s3d/S3D_Retracts_Closeup_1.mp4
Looking closer at the print you can see the first process using 0.3mm oozes some material which gets stuck to the outside walls
when travelling between the boxes. Also the skirts aren't really good, there's material missing/stuck in places where it doesn't belong:
www.aunix.de/tmp/s3d/IMG_1369.jpg
Process 2/0.6mm is starting to look better, but still there's some spots of material left on the outside walls. With 0.9mm they're gone.
And while it doesn't change on the outside any longer there's a difference at the start point/inner wall:
www.aunix.de/tmp/s3d/IMG_1409.jpg
(If you have a look at the lower right corner in this picture you can see an aspect of what jshell mentioned. Depending on settings
these corners might overshoot a bit or get bevelled/foreshortened. But I wouldn't start diving into this before everything else is
working.)
There's clearly some material missing - this is at 1.2mm retract distance, a bit too much for this direct drive/e3d v6 one.
This doesn't really give you values you can use because of all the differences, but maybe it can give you an idea of how to
pinpoint problems when printing. Like said in the other thread: If you're only having troubles at the start points, maybe
using a small negative value for "Extra restart distance" may help you there.
Another thing to check might be how hard your extruder grips the filament. There's often a spring which can be adjusted. If that's
a bit too loose the extruder might slip a bit instead of pulling back the filament completely but forward it the full amount after the
retract. Too tight and your extruder might have problems feeding filament at higher feed rates because of the resistance.
Oh, and yet another thing: Temperatures. Too high and the material will tend to ooze more. But as far as I can remember your
temperatures were in the ~200ish range which (to me) doesn't look to high for PLA at 0.2mm layer height (the PLA I'm using is
rated at 215°C +/- 10°C, I use it at ~210-220).
What are your retract distances right now, did I miss them or didn't you mention them here for all the cubes?
Notes:
1. All the images didn't show up when using the img/url tags. All links had "forum.simplify3d.com" in front of them, don't know why
that's not working as expected right now...
2. I know there's quite a lot wrong with that printer - it's my "tinker bot" where I change things all the time, so don't care too much
about elephant foot or vibrations or anything like that....
