First post here, and relatively new to H2D, with some prior printer experience.
I am having some challenges getting small scale prints to come out using Bambu ABS on my H2D. Latest attempt using “0.08 Extra Fine @ BBL H2D”, 0.4mm nozzle, on a smooth PEI plate. (And I have done calibrations, and Vision Encoder.)
I am getting a lot of “curling” on the underside, and around where the tree supports attach, above a brim.
Hopefully attached pics will help show the issue. Overall quality has issues too. 3mf attached for settings I used this time. Previous failures were similar, and I have run out of ideas to improve the results.
Small parts = short layer times = plastic isn’t cooling off enough on layer “N” before “N+1" is underway. The nozzle extrudes over plastic that’s still molten, so it “splooges out” under pressure. Slow down, turn part fan on, reduce extrusion temp 5-10º if the filament will allow it. And since these are small, the defect areas are in close proximity to the build plate, so lowering that temp a bit might help, too.
Thank you both for the suggestions. Slowing down a bit helped, but looks like I need to slow down even more.
I cannot control fan speeds, which according to what I’ve read was removed in BS 2.0.3.54 (I’m on 2.5.0.66), but that idea, combined with the suggestions to reduce filament and bed temps gave me a couple more ideas to try.
Looking at the latest attempt, I can see that the issues are worst on the side closest to the build plate. That confirms the heat suspicions. Furthermore you can see in the two images that the bearing pocket facing down, towards the build plate, has w-a-y more issues than the one on the opposite side which is facing up.
As much as the engineer in me hates to change two variables at a time… I’m going to try reducing the speeds even further, and placing the print on top of a raft, to get it both up off the bed directly, and away from any excess heat directly coming off the bed affecting the bottom side.
@Gerrit I was originally trying to print this with a 0.2mm nozzle, but the results were terrible, so I reverted to “0.8mm layer fine” with a 0.4mm nozzle to see if I could get things under control. Once I have these issues sorted out, I would like to go back to the original goal of printing with the 0.2mm nozzle, for the added detail.
Above settings can help to get a bit more cooling on overhangs (the parts facing the build plate), but also lowering the minimum feed rate, enables longer layer times, and thus more cooling.
Ah, thanks. I’m too new with BS to have seen those settings buried in the filament profile! I still feel like I’m fumbling around in the dark sometimes.
I have a print running now to see what happens. In addition to slowing the speeds down again, I added a raft, to get the parts off the bed, and added some thin 3-layer “heat shields” (flattened & merged cube primitives) that are around the base of the parts, just to see if they buffer the heat coming up from the bed a bit.
I had a spaghetti failure and aborted. That attempt was clearly no better than earlier attempts, and a curling piece of ABS clearly snagged on the nozzle and yanked a part off the raft.
I changed to the “0.12mm Fine BBL H2D” profile, and decreased speeds even more. I confirmed fan speeds matched the suggestion above, and this is what it looks like now.
Any other suggestions or ideas? I don’t feel like I’m making any headway here.
I wonder if reducing the speed parameters is really working. That means a lot of parameters. You might consider to start with reducing the max volumetric speed.
Did you adjust the cooling parameters, minimum speed and layer times already?
Too lazy to read through all the posts. Has someone already asked the “did you dry it?” question?
Moisture in the filament causes over-extrusion. The slower you go, paradoxically, the greater an effect it will have. So you could be fixing one problem while making another worse.
ABS likes to absorb moisture from the air. If you haven’t dried it (and I would do 24 hours at 70ºC), even if you removed it new from its BBL packaging, you need to dry it as your next step.
It shouldn’t be this hard. I print ABS and ASA (and Nylon which is also challenging to print well) all the time, mostly default settings (non BBL filament), only occasionally have had an issue. Problems always related to insufficient cooling. But I always dry my filament.
Yeah, thanks. I probably should have opened with the… “Yes, I dried the filament, which is in a dedicated AMS HT…” 8h at 80°C, <8%.
And no, it shouldn’t be this hard.
I have tried unsuccessfully to find some methodical, step-by-step filament calibration process, where I can print and incrementally change a setting or two at a time, e.g. not 10 temperature, speed and fan settings, so that I can actually determine which settings contribute to which fixes / outcomes. All the current attempts feel too random, and I’m not seeing improvements.
I think I’m going to go back and just try printing some basic part, and see if I can get anything to print cleanly, then start over again with vanilla parameters / profiles.
Here are the test swatches I printed, 0.2mm nozzle on the left, 0.4mm nozzle on the right, before I started heading down this path. There was some curling on the 0.2mm on the overhangs, but none on the 0.4mm nozzle, although that was using the 0.20mm layer profile.
By “adjust” do you have some suggested values? The cooling profile values in my filament settings matched what you posed above, so I didn’t reduce them further.
Reduce the max volumetric speed to what? The Bambu profile is at 20 mm^3/s. Do you have a guess / suggestion? Or better yet a way to test what it should be?
No, I didn’t suggest any values, but merely pointed out the parameters to tune.
I don’t know which brand you are using, but you might want to go through the entire calibration process to tune your filament. A lot of the parameters seem to be way off
Some of the suggestions given haven’t worked out so far.
Some flow calibration and temp tower should already get you in the good direction.
A simple test with default 0.2mm layer height to see what the result is, would also help.
Seems a lot of the focus is on the small parts, but maybe much more is off.
It’s not impossible you just have a bad roll of plastic. Overnight yourself some ABS from a non-Bambu source like MatterHackers.
But also, I’ve pretty much switched over to ASA. Whatever ABS I still have is just burning off old stock. ASA is easier to print than ABS and it’s mostly the same, mechanically. Consider ASA instead?
Thank you both. I have ASA and other brands of ABS to try. The ksr FDM model seemed ok-ish, at 0.20mm layer presets, and I didn’t think I’d have this much trouble. The fact that it printed from this spool of Bambu ABS made me think things were going to be ok.
I’ll try printing that again with one of the finer profiles and see how it comes out. That might give me some clues and a better idea where to start tuning. I’ll add a Voron cube too, just for giggles.
Here are pics of the ksr printed at 0.20mm layer height…
This is the same issue as you’re fighting on your little knob-like things.
Flow too high? Temperature too high? Looks like flow to me.
Did we already talk about manual flow calibration and a temp tower? I know the whole point of BBL filament is that you don’t need to do these things, but…
Or if you’re an empiricist like me, reduce flow by a few percent in the filament profile and drop the extrusion temp by 10ºC while you’re there, and see what happens…
Thanks, that was the area I was paying attention to as well.
I haven’t found a good step-by-step manual flow or temp correction procedure detailed anywhere. If you have a pointer I would truly appreciate that!
Before I saw your reply, I had prepared a plate with the Bambu / Voron cube, but before I hit print I decided to add one of my cutterhead parts, and just “paint the heck out of it” with supports, just to see what would happen. I’m thinking much better support may solve 85% of the problem.
I’ll probably still fiddle with flow and extrusion as you suggest, but based on the cube, and other tests, I don’t think those are way off. Here is my cube and a part from last night’s test print…
On vibes alone I would decrease nozzle temp by at least 5C, and really dive into the preview panel to inspect wtf the supports are doing and where they are making contact, and maybe select a different support archetype. For this object I would select Normal, adjust the spacing of the interface to 0, and maybe try to keep the top Z gap and XY gap a little smaller than default, just so it can print cleaner with the ultra low layer height. may be hard to remove
I dont know what the models objective is but Im guessing that the ragged ring-shaped side wants to be flat, and you might consider giving up the robustness of a tilted part for the precision certainty of a flat plane. or flip it so that the other surface gets crappy.
Thanks for the suggestions. I definitely need to try some changes with the supports and your ideas, and as you said, look closely at the preview.
The model is a cutterhead for ornamental turning, that has two bearings, an internal spacer, and a shoulder bolt as an axle. Yes, the bearing pockets do need to be round and flat. …and I’ll eventually work my way back to the 0.2mm nozzle.
I’m revisiting this issue, after pausing it for a few days…
I am printing a different part, same ABS spool, dry and with the 0.4mm nozzle, I am seeing the culprit start to emerge and thought maybe I can use this to ask and solve the prior issue.
Much like I saw in the test swatches above, I am starting to see some “curling” occur on a thin edge that is sort of hanging out in space. In the right side of the attached photo, you can see the edge starting to curl up. Could this be avoided with support or ??
Photo and slice shown, both at layer 60, and the complete model on the build plate (4 copies).
Ideas? Feedback?
Thanks
P.S. My support material for ABS just arrived, and I’ll be using that when I try to re-print the failed cutterheads above.
Try toggling the “Slow down for curled perimeters” from whatever current state you are printing with. I’ve found some materials like this setting to soak in fan during that time, but some materials absolutely hate this and end up just using the extra time and nozzle contact to curl up. Worth a shot
8h at 80°C, <8%.
…and I’ll eventually work my way back to the 0.2mm nozzle.
