Symptom: Line width varies 0.3-0.65mm (target 0.42mm)
- Occurs ONLY at sudden speed changes (10→200 mm/s)
- Consistent and repeatable
- All PLA filaments
- No mechanical clicking/grinding
- New nozzle
Pressure advance works perfectly for corners without speed changes in slicer. But when sudden speed change is introduced, severe over/under extrusion happens.
Flow rate is calibrated.
I know about extrusion rate smoothing, it helps, but not in every case. And anyway the wall width is fluctuates even with smallest smoothing values.
Have you manually calibrated pressure advance before?
Yes, I tried manual calibration, it looks a little better, but still, the wall is very thin.
And this is what happens if I set travel distance threshold to 1 instead of 0.
To be honest, I’m very disappointed with the print quality. And I’m even more upset that I can’t do anything about it. For example, I can’t read the firmware code and understand why this is so (or maybe even fix). It seems to me that this is definitely a flaw on the part of the firmware.
See I am not the only one that is crazy. I have talked to some other people on here about the print quality and how it must be firmware related, and some very strange artifacts that could only occur from firmware problems. Open source is always the best thing for a company to do, especially on 3D printing. And when someone is spending in some cases over $2000+ on a single printer, I expect to be able to fully mod it in any way needed, in order to keep it running past the company not supporting it or their eventual bankruptcy. I bet you will still see Enders kickin’ around in the far far future for this reason.
2 Likes
If upload isn’t blocked for you, you’ll get much higher quality advice if you upload your test model .3mf so other people can try it.
sudden_speed_change.3mf (25.9 KB)
I would be grateful if someone could check it for themselves.
UPD: better minimal example is below
1 Like
Thanks for posting this, and it’s interesting going through all of the changes, but I should have provided better instructions.
Starting with a default configuration and profile (for everything, printer, model, etc), reproduce the problem behavior with as few changes (preferably zero) to the default configuration as possible. Also note if it’s an orcaslicer or studio save and the version used (don’t say “current”, determine the actual numeric version).
Thanks, I tried to make a minimal example.
sudden_speed_change_minimal.3mf (25.9 KB)
Just in case, I’ll send screenshots with all the changes, there are only 5 of them. You can also increase max volumetric speed, but even with the standard 12 mm3/s (which limits speed to 159 mm/s), I get a hole.
Bambu studio version: 2.4.0.70
That is much better. Now are all of those (few) changes absolutely necessary to reproduce the problem?
I have had a look at the first 3mf you uploaded and there are quite a lot of puzzling changes.
- Denoting your PLA as “Bambu PLA Lite”, is that PLA Basic or PLA Aero?
- What are the Customizations in the Machine and Filament Start g-codes? There’s probably a very good reason for them, but if they affect the calibration, things could get out of control.
- Your test model is not actually comparing a speed transition.
You are comparing a modified scarf seam performance at 200mm/s to a modified scarf seam performance at 10mm/s (and many other changes such as a relatively high flow at a really, really low layer time) on two distinctly different objects.
If you want to actually assess a speed change rather than your scarf seam settings, use a single block with a modifier.
Also, if “PLA Lite” = “PLA Aero”, it easily explains
as well as major speed/flow affected differences.
I’d recommend to start fresh by resetting all changes back to defaults and then take it one step at a time, posting pictures, 3mf’s and further important information (exact filament and drying information) as well as ensuring that test pieces do not undercut critical parameters such as minimum layer time.
& 
1 Like
Bambu PLA Lite is a value version not available in the US. I have some here that someone brought over from Asia for me, and you can order it online. It’s a cheaper version of PLA Basic, it’s not special or Aero. It’s not good/better in any sense, it’s just more “basic” than even “PLA Basic”.
1 Like
No repro on P1P with your 2nd 3mf as is, I’ll try some of the H2 class machines next.
This is one of those models where prusa’s newer preview model showing “actual” speeds (aka after accel/je rk and starting velocity are factored into cornering) in regions rather than visualizing the prescribed speeds, would help reveal details. You can half-■■■ this in orca/bambu slicer by putting “extrusion rate smoothing” in, [only to preview - if you print like this you invalidate your tests].
Perhaps also a hint: the auto arrange wants these objects further apart, a seam right next to a seam with 0.32mm travel is going to be essentially a retraction torture test.
edit: wtf je rk is a censored word? It’s a literal setting
Thank you guys.
That is much better. Now are all of those (few) changes absolutely necessary to reproduce the problem?
I haven’t been able to do it any other way yet. But I think it is possible to make printed parts bigger and remove the need in “disabled slowing down for better layer cooling” and remove “top shell layers” override.
Denoting your PLA as “Bambu PLA Lite”, is that PLA Basic or PLA Aero?
As already noted, it is cheap version of PLA basic. However, I was able to reproduce this problem with PLA Matte.
What are the Customizations in the Machine and Filament Start g-codes?
Initially, I used OrcaSlicer with machine and filament g-codes taken from bambu lab studio,
because they are newer. Idk, why Bambu Lab decided that they have changed.
You are comparing a modified scarf seam performance at 200mm/s to a modified scarf seam performance at 10mm/s (and many other changes such as a relatively high flow at a really, really low layer time) on two distinctly different objects.
…
If you want to actually assess a speed change rather than your scarf seam settings, use a single block with a modifier.
I tried to do this way, it actually makes seems even closer. I couldn’t get continuous line with speed change in it.
1 Like
Not sure what you mean, these speed changes aren’t “close”, it prints one object layer at one speed, then one object layer at the next.
I mean that in the initial model I left a small gap between two objects, while the model with modifier does not have any gap between walls with seems.
Very interestingly, I rechecked myself again with the second 3mf and got the same bad result. Maybe it is related to P2S or to specific hardware version. Btw, I have 01.01.01.00 version of hardware.
PLA Lite Cyan on the top, PLA Matte Ash Grey on the bottom.
