James M. Woodward

Alright, Maybe I take it back.

I believe disabling z-hop helped, but using #orcaslicer I cannot figure out why I'm getting little bumps on the threads.

#3dprinting community, any ideas?

This isn't any kind of axial wobble, doesn't matter if I print fast or slow, cooling high or low (though I don't know why low would be good), arc fitting, precise walls/height, at 0.3 height or 0.1 happens, print temps high or low (this is PLA so, #easymode), wall print order, etc. All of it doesn't change this and the rest of the part itself is perfectly smooth. This only affects this NPT threaded portion.

I plan on using a die (as in a tap & die) for final assembly but I would still rather know WHY it's happening.

Jun 29, 2024 at 6:52pmWeb
Show threadJames M. WoodwardJun 29, 2024

Tried 6 walls with 100% infill, no change.

The problem is hard to take good photos of, I may try a different filament just so that I can see what's actually going on better. I don't suspect this is a filament specific problem.

Show threadcybervegan (moved)Jun 29, 2024
@jmw It's possible that your filament is damp - it looks like steam bubbles. Do you hear a popping sound during printing?
Show threadJames M. WoodwardJun 29, 2024

@cybervegan
We are certainly at the high point of our yearly humidity here but I live in the dessert. I leave PLA out for months without issue.

To answer your question however; No popping/steaming visible and if it were steam; That would be an issue on the outer walls everywhere, right? Not just the threads?

Show threadcybervegan (moved)Jun 29, 2024
@jmw Yes, you wouldn't expect it to be just in one place. What do those areas look like in your slicer? One thing with threads is the *overhang* combined with the layer stack being significantly skewed each increment. I'm now wondering if it might be an artifact of the way the slicer is trying to deal with that situation?
Show threadJames M. WoodwardJun 29, 2024

@cybervegan That's my leading theory at the moment since it's (literally) an edge case. However, I tried supports and got the same artifacts. I just can't quite put into words why it's doing what it's doing. Curling, under/over-extruding, etc. makes sense, but 'zits' are a bit strange. Unfortunately FDM being FDM I don't have another way to print this. I could try splitting it but it's hollow part so 🤷‍♂️

I tried Cura but it's been a LONG time since I used it, and it tried to 'correct' the threads somehow, and made them thread-like.

It's not a fix but maybe I'll crank the extrusion down and see if that fixes it. Maybe this is a small part/over-extrusion problem.

@silvenga The photos from the slicer itself are included in the last post. I see nothing in the gcode/slicer that indicates a cause.

I've tried aligned, shortest, etc. No change on any of that. All of the sides roughly look the same, and even with it crawling along at 50mm/s and 100% fan at 195C the artifacts were still present. No blobbing on the nozzle, I clean it between prints.

Show threadSilvengaJun 29, 2024
@jmw @cybervegan hmm... How dialed is your retraction/preasure advance? Could cause fluctuations in extrusion.
Show threadJames M. WoodwardJun 29, 2024

@silvenga @cybervegan

I'd say 'it's not' really. There's no pressure advance in play (DD w/PLA).

I dialed it in enough to prevent stringing, and otherwise have done squat with it, but I don't believe there are retractions in those locations. (see photos from previous, previous toot)

Show threadSilvengaJun 29, 2024
@jmw @cybervegan if retraction is too high you may see over/underextrusion as the drive loses control of the filament (eg retracting so much the filament "breaks away" from the molten plastic in the nozzle). Similar with incorrect preasure advance - I've seen PA impact perimeters 10-20 mm from the detraction point - it almost looks delayed.
Show threadJames M. WoodwardJun 29, 2024
@silvenga @cybervegan
While true, using direct drive my retraction is 0.4mm, not quite applicable.
Show threadSilvengaJun 29, 2024

@jmw @cybervegan yeah, just rubber ducking here. I know orca slicer's default retraction is quite high - for some reason I'm remembering 2mm? I know prusa slicer defaults to 0.8mm - which can be too high for some pla. (all DD)

Ocra has a really fast PA calibration - might be worth just doing it. I really wish prusa slicer would add filament overrides for PA, Orca makes it trivial. Prusa hard codes PA into the gcode, blah.

Show threadJames M. WoodwardJun 29, 2024

@silvenga @cybervegan

I have PA disabled in the printer as the testing I had done with it previously showed no benefits at such small values. I think the OOB retraction in Orca was only 0.8 or 0.6.

Show threadFritz AdalisJun 29, 2024
@jmw
It's weird how the bumps are pretty regular. If you look at the file in a gcode viewer is anything happening there like an un/retract?
Show threadJames M. WoodwardJun 29, 2024

@FritzAdalis Not in the slicer, but let's see.

Attached are four photos;
1) OrcaSlicer on Layer 36, near the end of the layer.
2) Zupfe GCode Viewer - Layer 36 - Top Down
3) Zupfe GCode Viewer - Layer 36 - Side
4) "GCODE ANALYZER" - Layer 36 - Top Down

By #1-#3, everything looks kosher. #4 looks extremely jerky but I don't know that it's a great tool.

But if you look at #Orca / #Zupfe, they both show multiple circular walls, none with retraction/etc. in them.

Show threadFritz AdalisJun 29, 2024
@jmw
Yeah, all nice circles. Hm.
Show threadJames M. WoodwardJun 29, 2024

@FritzAdalis I just stopped a test print with a black filament (see attached photos).

Photo 1 - Shows that this isn't an issue except on the threads. The test print here looks great.
Photo 2 - Attempts to diagram what's happening in the Z plane with the 3d printer. This is very pixel-peeping. It can be hard to see, but zoom in and follow a filament strand up and around and you'll see what I'm talking about.

Here's what I think is happening. This is 3/4" NPT, there isn't enough overlap with the layer below each to support the next so heat/cooling/dropping and curling are all important.

It's also important to remember that it's a coil/thread. That means that it's going to climb through the Z layers, meaning that the lines are going to wrap around each other BUT they're planar so the printer cannot 'vase mode' the spiral, it's continuously chopping and then restarting the coil on the next z-layer.

I think @silvenga might be onto something with retraction in such an edge case/unsupported condition that I'm getting zits, but I think part of this is simply what happens when you start printing coils that climb through Z layers on FDM.

Show threadSilvengaJun 29, 2024

@jmw I'm assuming you have aligned seams enabled, I'm also assuming different sides look the same (suggesting it's not a heat issue of going too slow, causing sagging).

Is filament blobbing up on the nozzle? It might be leaving filament behind. I would also check the slicer gcode preview, it might be generating something odd, causing incorrect overhangs.