This looks to me like a pressure advance and/or acceleration issue. The top of the hole layer is printed at overhang wall speed (probably 20mm/s) and then it shoots up to outer wall speed which is probably 200mm/s.
Looks like pressure advance isn’t coping leaving a gap and then a blob when it catches up.
I would try increasing overhang wall speed (no point slowing down for such short overhangs although it may mess up longer overhangs elsewhere). I would also try reducing outer wall acceleration values. Upping the minimum print speed in the filament profile could be another option. Not sure if some of those settings are only available in the softfever fork I don’t use the stock studio.
I would guess PLA doesn’t show the problem simply because it needs less pressure advance.
later: Should have said the pressure advance K value being correct is of course important.
even later: Extruder jerk setting in the printer profile seems low. Maybe something else worth trying.
Yes, I know that it probably has to do something with pressure advance. I tried many things. I printed some files Bambulab sent me. I also determined the K value manually:
I expect a solution from Bambulab. I just don’t know why only few people seem to be affected. Maybe I should do a poll to find out.
They occur when the head is accelerating after slowing for a wall overhang. The extruder has to accelerate at least in advance and possibly even faster (I never looked into exactly what pressure or linear advance algorithms do). If the extruder can’t keep up with the head you will get a hole and then when it catches up an over extrusion blob.
That is my theory which matches what you see. If you slow down head acceleration you give the extruder more chance of keeping up.
It should probably be fixed in printer firmware. When pressure advance requires the extruder to accelerate faster than it can head acceleration should be limited to match the extruder’s ability.
It depends how much it slows down. The top of the hole and only the top of the hole is considered to be 100% overhang and it slows to probably the minimum limit in the filament profile. Look at line types and speed in the studio preview.
The print you just posted looks generally horrible is your filament dry?
On the left is a print in eSUN PETG I did when this thread started not sure what profile and I noticed the ripple to the right of the big hole and the tearing top right of the other two holes. I thought yes that is a mild version of the same problem, strange, and didn’t bother thinking more about it. Today I revisited the thread after seeing it drag on for so long and did think about it. The print on the right is the same PETG, 0.2mm standard profile with outer wall acceleration reduced to 1000. Not sure if I calibrated in the mean time, but, today the PA K value for the filament is 0.032.
Maybe my PETG is no longer dry. But I had it in the oven for 8h @80°C, so it certainly was dry. If you scroll up you will also find some images I posted after drying.
Your print shows exactly the same issue. Yes, it is a little less pronounced, but otherwise it’s exactly the same. Maybe if you add up all positive effects such as drying, increasing temperature (up to 280°C), reducing speed and acceleration, then it looks better. But this is only fighting symptoms and not addressing the root cause… One main point in buying the X1 was because I wanted to print PETG. Now I had the printer for 6 Weeks, and I printed probably more than 1kg for trouble shooting. So far I can only use it for PLA.
Did you open a support ticket? Maybe things speed up a little if more people complain!
The print on the right was done today with 0.4mm nozzle and the only change to the 0.2mm standard profile being outer wall acceleration reduced from 5000 to 1000. Nozzle temperature was 235C. It shows zero sign of ripple or holes to the right of the large hole and no tearing at all at the top of the other two holes (or any of the holes on the other sides).
It looks like the holes bulge out on the right, but, it mostly just looks that way because the reduced acceleration and so print speed leaves the surface to the right of the holes more shiny. If I wanted pretty I would slow down outer wall speed.
I’m pretty sure you’ve got an extrusion issue in your filament path. I regularly print with PETG on my X1C and do not see your issues. I had problems with overhangs in the Arachne engine, and I do tune my speeds down ever so slightly. I tend to keep the speeds the same inner outer etc, just lower for support, but that “delamination” line points to something interfering with the filament flow. In another printer I have, it was a small piece of broken filament in the extruder gears. The varying speeds of the extruder cause this to move around, and I saw this in my random print artefacts too. I now use the SoftFever fork of Bambu Studio as I’m trying to dial in the Arachne engine profiles. I’ve also read people have issues with the AMS feeding (I did and had to pull out some ground up broken filament from my extruder #2-feed errors & broken piece left in #4- old filament), so some have used the rear spool holder to eliminate this for testing. Worst case you might have to inspect the DD printhead extruder which is a real faff just to eliminate this potential fault.
For me this would not make sense because the artefacts in my case are not random. Further it would not explain why PLA shows no issues. The AMS is certainly not the problem because the same issues are occuring when printing without.
PETG printing at 255 with standard P1P Generic PETG on textured plate.
I have a test cube example with print quality issues which may be the same one. On the front X-axis, printing from the left Y-axis there are linear protrusions, not voids, that extend from the "v’ of the dimple pattern to the left and similar protrusions on the post at the right side. In the photo I’m holding the cube vertically as printed.
This issue of protrusions is not nearly as evident on smooth surface prints with curves or corners.
There are other quality issues with this print but this is the main one and it may relate to this discussion.
The remaining 4 surfaces are almost perfect.
Thoughts?
WildRoseBuilds Test Cube: Printables
