iverstaylot00
-
Posts
6 -
Joined
-
Last visited
Content Type
Forums
Events
3D Prints
Posts posted by iverstaylot00
-
-
Hi all! Sorry for the late reply, I got absolutely swamped with labwork for midterms.
@burtoogle your custom build worked beautifully, and as per your suggestion I disabled wall overlap compensation in the main cura release. However, your anti-resonance feature has me hooked! Probably will be sticking with your build!
@gr5 I cleaned out my z screw and and applied fresh ptfe. Found a tiny bit of dirt that may have been a bit of a culprit in the over/under extrusion. I also tried out uniform wall extrusion speeds, and I noticed an improvement in the zseam! Also, I have a slight suspicion youve been the mod approving my posts, so thanks for that as well
Thank you both for your suggestions, I can post my test print results if you would like, however I think you can imagine what a creality dog looks like
Tl;dr to anyone reading this who had similar issues, try disabling wall compensation, or try out burtoogle's release
-
9 hours ago, gr5 said:
Oh and to explain what Z screw has to do with this - you are doing 0.2mm layer height. This will be incredibly accurate on average but each layer may be a little lower or higher than 0.2mm. If some are 0.19 (doesn't move enough due to sticktion) and then one is suddenly 0.21 then the 0.21 will appear underextruded (not stick out as far - a mini valley).
Whenever the distance between nozzle and bed move less than 0.2 it will be slightly over extruded and stick out and the other's will be slightly underextruded and stick in giving you these horizontal lines.
This effect is MUCH worse if you switch to 0.1mm layer height.
This effect tends to be pretty repeatable as the dirt on the Z screw doesn't typically move around much print to print. So you can print a Cube and then a pyramid and typically they will have mostly the same horizontal lines in the same spots.
Wouldn't a simple change between slicers excuse away the dirt problem? And also, I clean my machine very often so I doubt that's the issue. However, the fact I print my walls in 2 different speeds may be causing over/under extrusion definitely peaks my interest. I'll try out printing some test prints at a uniform inner/outer wall speed after I'm done trying out burtoogle's release and see if that helps. The thing I find odd though, is that I never had this issue until I updated to 4.5. Which confuses me even more. I've been printing with that profile for about 6 months now
10 hours ago, burtoogle said:BTW, using small values of max deviation with UM cura releases can give rough results. You may get better results using one of my releases that can be found at https://www.dropbox.com/sh/s43vqzmi4d2bqe2/AAADdYdSu9iwcKa0Knqgurm4a?dl=0
Please read the README.md file there for more info.
Also, you may want to disable the overlap compensation as that tends to cause artifacts.
I've downloaded your release, running a test print now! Thanks again for your help 🙂
-
Just now, burtoogle said:
BTW, using small values of max deviation with UM cura releases can give rough results. You may get better results using one of my releases that can be found at https://www.dropbox.com/sh/s43vqzmi4d2bqe2/AAADdYdSu9iwcKa0Knqgurm4a?dl=0
Please read the README.md file there for more info.
Also, you may want to disable the overlap compensation as that tends to cause artifacts.
Thank you so much for your help! I will try this after I get home from class and inform you how it goes!
-
38 minutes ago, burtoogle said:
Hi, I have looked at the gcode in a viewer and it doesn't look lumpy-bumpy so I am guessing that the issue is not due to the gcode having poor coordinates (i.e. the corners are where they should be). UM Cura releases do have some bugs in the code that simplifies the wall polygons (the code that uses max resolution and max deviation) but I don't think that is causing your problem.
What I think is more likely is that the settings that you are using combined with the path that Cura generates is causing the print head to "wobble", i.e. it's a mechanical issue brought on by how Cura is driving the printer.
Do you have some sample gcode from another slicer for this part that doesn't produce the same variations in line positions?
I do not have some comparison GCODE for that particular part, however I do have a pair of the Creality dogs! (I don't feel like reprinting a 30 hour print just to test, lol)
dog-2.5H is done with S3D, and has none of the seen inaccuracies (however the zseam is not configured properly, ignore that. I do not own S3D, a friend does and sent this to me)
cura-dog is sliced with my profile. They both have identical layer heights and very similar speeds, and other things. The only other major difference besides the zseam is the fact that the S3D model prints with a raft (which does not solve my issue, I have tried it but forgot to mention it)
I can get some photos for comparison if you'd like in about 8 hours, as I just left for uni and won't be back until dinner
-
Hi there, although this is only my first post on these forums, I have been using cura for many years for schooling and personal hobby. Recently, in the 4.5 update to cura, I have had issues reminiscent of the other forum topic I linked below from 2014. After a lot of testing on my end and researching other topics, I have narrowed these issues down without an inch of doubt to cura, as no other slicer I tried has this inaccuracy. Attached gcode of one of the models I printed that has this issue. I've attached a picture of the printed version of a similar model. Please note, it is NOT because of any machine flaws like loose belts or the sorts, I have near-perfect walls on other prints.
Some Experimenting I've already tried to no avail:
Lowering travel to as low as 80mm/s
Lowering max deviation from the stock 0.05 to 0.035 (it's a multiple of my steps per mm)
Lowering print speed & accel/jerk
Different layer heights
Many printing orientations
Drastically different print profiles to make sure it wasn't one or two particular settings causing this
Tightening belts, lubing parts, (just to triple check no machine errors)
Some settings that may interest those who are way smarter than I am:
Layer Height: 0.2mm
Max Deviation: 0.035mm
Printed on: Ender 5 with TL-Smoothers & PTFE bowden
Accel: 350mm/s/s
Jerk: 6
Base Speed: 70mm/s
Outer Walls Printed @ 35mm/s
Inner Walls Printed @ 52.5mm/s
Travel Speed: 150mm/s
I know these speed settings are VERY conservative for an ender 5, however I like my print quality. Also, I hope this topic doesn't rub off as rude. I absolutely adore cura, and that's the reason I've been using it for years. I assume I am not the only person having these issues, but after a lot of googling I oddly enough wasn't able to find another topic like this other than the one below. If I've forgotten to attach anything important, feel free to ask. I tried to include everything relevant (and then some)Also, as a last note, I know the attached image is not of the same model I submitted the gcode as, however it's sliced with EXACTLY the same print profile. I have the printed version of the submitted gcode, however my phone really struggles to capture the errors. The errors are easily visible on this portion of the model, the flat trapezoid shaped area really brings it out. The forum topic from 2014 I linked has some wonderful representations of the gcode causing these errors.
Does anyone know what this banding is?
in Improve your 3D prints
Posted · Edited by iverstaylot00
Hi guys! I'll try to make this as detailed as possible to help anyone who is kind enough to provide some insight 🙂 I hope this is not too much information, however I want to make everyone's job easier by providing all needed info right off the bat. Also, I don't mean this to bash cura. I absolutely adore it. Built-in tree support generation in particular has me hooked.
For a while now, I've been getting these weird print artifacts for a single layer. It only happens along layers where there is some sort of major feature change (like the end of that joining material between the two towers). I've also seen it happen with many other features. In the benchy model posted down below, I added some colorcoded lines to show off other features that can produce this result. I'm 99% sure this is not a machine issue, as it's reflected in the slicer. Plus prints in other slicers turn out fine. Nevertheless, I've checked my belt tension, calibrated esteps again, and entirely replaced my zscrew assembly (for a lead with one start, not four). None of these things have fixed the issue. I have also tried slicing on a separate computer with an identical profile (to make sure it's not a corrupted install), and I am able to reproduce the same issues. I also found this post on imgur from a user who is having an extremely similar issue to mine https://imgur.com/gallery/P6HFD, notice how his banding is in very similar areas as mine.
I have provided the sliced gcode I used to print for both the caliper mount and the benchy! I have also provided a download for my cura profile I used to slice these models down below (along with screenshots for a faster glance at my printing speeds)
banded-benchy.gcode
Caliper Mount.gcode
Caliper Mount.stl
Print Settings w/ Custom Overrides (print specific)
Printed on an Ender 5 with TMC2209s. I swapped in the stock mainboard with A4988s for the hell of it to double check it wasn't an issue with my MB, and reproduced an identical print using the same gcode.
I am not 100% sure if it will reflect this in the profile download that I gave, but I manually changed the layer height from 0.2 to 0.12mm (which does not require any microstepping by my z stepper)
i-used-this-profile.curaprofile
Model Specific Overrides for both Benchy and the caliper mount
Model Banding
Going from the top to the bottom for the benchy;
There are some other minor banding artifacts that I did not highlight that are caused by the top layers of the crate in back, top layer of the little cylinder in back, and top layer of the steering wheel and dash (I know its not a car, but thats the only term I can think for it)
Am I expecting too much from cura? I dont mean to sound like a dong with this, I mean it entirely constructively. It's a slicer developed that is free for all to use.
Again, thanks in advance for any and all help given, I hate to ask others for help but I've tried so much self-experimenting and been unable to smash this problem. It's safe to say this has been frustrating me quite a bit, especially when I contours in vase mode that are so flawless they reflect the individual facets of the original STL (photo below)
EDIT: I totally forgot to add, for some reason I wasn't able to replicate these bands when checking the layer view in an external gcode viewer. Perhaps that is because cura represents extrusion widths in layer mode more accurately than said external viewers?