I’m not really sure how to explain this, or what it is called to even start searching for remedies. I have opened a ticket with support, we’ll see how that goes. Essentially, vertical walls look like veritcal vfa. If there is a pattern in the vertical wall print, it creates almost layer shifts where the patterns are. Sometimes this is minimal, but other times it honestly makes the wall surface feel like it has ridges and looks that way too. I am posting pics, and a link to the model.
Printed in Bambu PETG-HF Black, but PLA and ABS have also had the issues. This does show up on my X1Cs and P1S, P2S printers as well, but nowhere near as extreme as it doesn on the H2D. This is something I have been meaning to address for since getting the printer on release, but time is a thing…
When I had this problem it turned out to be that I did not clasp the nozzle correctly. It caused similar artifacts to what you are showing. If the nozzle is clasped correctly, then I agree the other thing I can think of is Benchy hull line. In the slicer preview tab, change the view mode to layer time/speed and see if that lines up with your artifacts
Pretty much this happens with any print that has any sort of verticle design or hole in the wall. I don’t really accept this is the way it is. The prints are pretty awful.
This is what I know. I have 7 Bambu printers, and the most expensive one I have doesn’t get used for my business because it doesn’t print as well as my X1Cs or P1Ss.
Regardless of what you accept, you have to learn why it’s happening. It’s clearly benchy hullline, so you change slicer settings to reduce the effects, it’s that simple.
The reason benchy hullline is tricky is that to solve the problem you almost always have to sacrifice something. This is physics, you can’t bypass it. Not having the same level of artefact on other printers simply means that they have sacrificed other things knowingly or unknowingly.
Why is it benchy hull line? The hull line on the benchy happens where you get the solid layer from the deck. This at least makes some sense with the walls being printed with some material pushing into it. Although this is a setting I generally try to minimize and reduce that bulge that happens from solid layers.
In my case, there is nothing there to push the filament out vs any layer beneath it. If the layer speed is the same, cooling should be the same, why should one layer protrude more than another?
And, a question? Why should anyone accept this? Bambu is supposed to be essentially plug and play with great results. This is after all why I have so many of their printers vs another company.
If someone has an H2D, I ask that the print one if the bins from that site and post pictures.
Benchy hull line happens whenever there’s uneven Layer time. Which is clearly shown with your slicing preview by switching to Layer Time view.
In your case, it’s more than just benchy hull line, but also your filament choice of shiney petg. The longer the filament stays in the nozzle, the more sufficiently it gets melted, and the more shiney it becomes. The shine changes depending on the layer time too.
Plug and play doesn’t mean that you would get the look you want. This part is functional as is, just no good looking. There’re already tons of this kind of discussion previously and it all turned out to be just benchy hull line.
There’s no “one profile fits all” for FDM, if you believe so then unfortunately it’s a lie.
Bambu support recommends printing outer walls first and slow down the speed.
I still don’t understand why my other printers print this part better. I looked at the layer time slicing it on an X1C vs the H2D, and it looked the same, yet the X1C prints it better.
I am honestly confused. Yes, the explanation makes sense, but so far there has not been a good explanation of why it prints better on the cheaper printer. Honestly, people kind of get hostile about anyone saying their cheaper printer prints better.
And yes, I get it, 3d printing is not perfect. I have to explain this all the time to people I make parts for. Because let me tell you, the people buying stuff from you, want perfect.
because the fan is different and the nozzle design is different and the hotend design is different. also there’s no chamber cooling for older printers.
a stronger fan = larger cooling difference.
a faster hotend = usually longer melt zone = larger shine difference.
chamber cooling = less clogging but higher chance of over cooling
This overhang requires 100% fan cooling. So if you have Pre start fan time at 2s, that means the entire outter wall would be printed at 100% fan for this geometry:
now i’m starting to think maybe it’s the slicer’s fault, the default profile’s 2second pre start fan time is obviously causing problems for challenging geometries. I’ve submitted an issue on github to urge them to turn this off by default or give us better options than current implementation.
I will try messing with the cooling, thanks.
But as an fyi, it isn’t the shiny difference that is the issue. Sure that doesn’t look good. It is actually the texture it gives the walls of the print. It is for lack of a better word, ribbed.
These storage drawers are for me, I stumbled across them and thought I could use something like that. So the roughness of the print isn’t a huge concern, but the same type of imperfections happen on some of the prints for my products. I figure if I can solve it for this, then maybe it will work on my parts as well.
It’s better in some cases, but still creating bad results with other challenging geometries.
The core issue here is that the fan mover in gcode can only work with line segments. And line segments can be very long (like this outer wall).
So basically it’s not a “easily perfectly soluable” problem with the gcode centric FDM. Very unfortunate. But still for this part i think turning off pre start fan time would be better than leaving it on.
yes it’s ribbed because of the, over cooling. it’s blasting fan at 100% all the time for those textured areas. sorry this isn’t really your fault, but a problem that’s just very tacky, and it’s caused by another compromise that we’re facing…
So basically, fans cannot be boosted from 20% to 100% instantly. Ideally it should but in reality it takes 1-2s to actually do that. So to improve overhang performance, at some time Bambu Studio and Orca Slicer introduced a new algorithm to turn on the fan EARLY, so that it ensures the overhang quality.
This usually won’t cause any problem for PLA, but it would become a problem for filaments like PETG. Especially PETG, because its overhang fan speed is usually still kept at 100%. While higher temp filaments usually capped the fan speed much lower, for overhangs.
So now the problem is your walls are printed with 100% fan when it only needs 20%.
Is bambu doing something wrong? no. it’s needed for the overhangs of the petg.
Is bambu doing something right? no. it’s overcooling the wall.
So why this is happening? because we’re printing gcodes. And gcodes can only operate the fan on a line to line basis. That means all the line segments within that “2 seconds” time window would be affected by the fan boost.
For example, if Line D is overhang, Line C takes 0.1s, LineB takes 0.5s, and Line A takes 3s, unfortunately Line A would be affected too, by the default 2s pre start time.
That is interesting that the H2C isn’t having issues. I want one, but am afraid to buy one since it is an H2. I am kind of curious how the H2S does with benchy hull line.
