I've been working on trying to start using some of the latest features in Marlin. A few weeks ago, I implemented Junction Deviation and thought I had a pretty good value of .03 with an acceleration of 600. Yesterday, I started working on linear advance. I printed several iterations of the test pattern and settled on a K setting of 1.35. This is with LA 1.5 and using TPU filament.
The printer is my redesigned i3 cartesian with a Titan direct drive extruder using 1.75mm filament.
I am printing a single wall test object that has 90 deg corners as well as rounded corners. It's the object located
here.
After configuring the junction deviation, the object printed mostly ok, except the spot where it was starting a new layer would end up with a pretty big blob of filament. That's when I went to attempting to work with linear advance. After setting the linear advance and unretract values that gave me what I would call almost perfect test patterns, I went back and printed the object above again. As you can see from the photos, the 90 deg corner barely has any filament and you can see right through it. Same with the entrance to the big rounded side. There are no retracts/unretracts happening at those locations and the wall widths are correct at ~0.45mm. So that begs my question, does that seem to be a problem with the junction deviation settings, or the linear advance settings?
This is using Marlin 2.0.5.3. The printer is using a MKS-Gen-L board with DRV8825 stepper drivers. I should also mention that I'm using Volumetric Extrusion. I did the linear advance test pattern both with VE turned on and off and didn't see any real difference.