Wanted to test out @Tom_Oyvind_Hogstad suggestion of replacing lines of Gcode based on Z values. I skimmed this model, which included some interior overhangs that’d make a hollow print impossible without issues. So, I compiled a 30% fill version and only replaced the gcode for the layers featuring the issue.
Although I was a little generous with the amount of layers, it came out pretty well outside of my flowrates not calibrated very well to my old makerbot filament. Due to this, too much plastic gradually raised the model to almost snapping half-way through. Oh well!
Enjoy.
http://www.youtube.com/watch?v=PmQk9iq_Wf8
How were you feeding filament? It looks like the spool was pulling. You need a feed tube and something to allow the spool to turn freely.
I fed it through a guide but I still gotta grab that tube, yes
I found some cheap HDPE tubing tat works nicely. Stop by the shop and we’ll cut you off a piece.
@Whosa_whatsis what do the artifacts typically look like when the spool is pulling?
Check out the last shot of the print in the video above. You can see it in the face.
I see what looks like somewhat irregular z-wobble, holes in the layers and certain layers being a bit thinner. Which one do you mean?
The holes are likely my retraction set a bit too high. I exaggerated up the filament diameter a bit, so I wasn’t getting the best flow rate by the top. My premium filament is a lot better calibrated than this but I’m glad I took that micro picture of the face!
Neither the period nor the amplitude of what you describe as looking like Z wobble is regular, and it’s the compression, not the position of the layers that are uneven. When the spool pulls, the extruder’s feed makes it climb the strand of filament and flex its mounting a bit, making those layers less compressed. When the tension is great enough to make the spool turn, the extruder will drop back to where it’s supposed to be and the next couple of layers will be over-compressed.
@Ben_Van_Den_Broeck This print demonstrates the issue really well. I’d like to get a picture of it to illustrate spool-resistance Z artifacts for my article.
@Ben_Van_Den_Broeck i’m in no way trying to sound condescending, by the way. I appreciate the amount of experimentation and research that going on in this community, even if quality isn’t always perfect - mine isn’t either.
@Whosa_whatsis thanks for clarifying that - I have seen similar artifacts on some of my prints as well, but never suspected the spool to be the reason. Guess I’ll have to get me a feed tube!
“Similar” is a relative term. All Z axis artifacts look similar if you don’t inspect them too closely, which is why I recently wrote 2000 words on the subject: https://plus.google.com/105535247347788377245/posts/4swesothSep
I’ve seen your article, but haven’t had time to read all the way through it.
Regarding Z-wobble, the Sells Mendel is the worst of all the machines i’ve built - you might want to correct that. It’s not all about how oftern the rods are constrained, but more about how centric the couplers are and how stiff the threaded rods are.
@Whosa_whatsis i’m pretty sure it is “real” wobble - i mean, after all i can see the hotend’s tip (and the threaded rods) wobbling around when Z is jogged. I think the problem here is that the 608 bearing has an exact 8mm bore while the M8 rod is a fair bit smaller on the outside, which makes it hard to affix the rod to the bearing at dead-center.
I’ll take a close look at the bearings the next time i’m there.