I configured my own Marlin for SKR Mini E3, but since moving to a Hemera I am getting worse results than with the stock Ender 3 extruder, and am wondering if I missed something or got something wrong, so was hoping maybe someone had a known good Marlin config for the aforementioned combo For reference, here is a link to the config I am using at the moment - I took the BigTreeTech fork of Marlin (Which is unfortunately not current) and applied changes for Hemera https://github.com/evilC/BIGTREETECH-SKR-mini-E3/commits/Hemera
Well I wasn't looking for help with the specific problems, just firmware, so that's why I did not go into detail on that (Believe me, as a FOSS developer, I too hate people saying "It doesn't work" with no real info) Seeing as you ask, I cannot get a single print to complete anything like reliably - I previously had a Basaraba single direct drive extruder, and if I print using a gcode file which printed fine using that (Same gcode, same material, same bed), I get all manner of problems - warping of the print, what I can only describe as "Inverse elephant's foot" (Print wider at top than bottom), and straight up weirdness such as the following picture, whereby suddenly part way through the print, one part of one layer is drastically higher (like by 1-2mm) than the rest of the layer: FYI there was support underneath that overhang However, there are so many potential variables at play, rather than seeking help addressing the specific problems, that's why I was hoping to get a Marlin config from someone with the same setup, so that I could eliminate incorrectly configured Marlin from the equation It's maybe also worth mentioning that the other thing that has changed is that I am now using a 5015 part cooling fan, along with a different duct than I was on the stock Ender 3 - this is my setup:
The image you show looks like a serious cooling issue. I doubt the same gcode you used for your previous extruder/fan combination will come close to working with a new fan duct and a Hemera. As an experiment, though, try using that gcode, but keep the fan kicked up to 100% after the first 2 layers (you may have to sit at the controls and keep kicking it up, since the gcode probably fiddles with it constantly). Much better would be to re-slice and have higher fan settings. gcode is anything but portable across a change of your machine setup.
The fan in the original gcode was set to 100% Here are the settings in Cura: I'm no expert at interpreting these settings, but it seems to me that the fan would come on full at layer 4 (Which IIRC was about the time I remember it coming on) which is way, way before when this problem happened - looking at the picture, I would hazard a guess that the issue starts happening around layer 15? Layer height is 0.2mm
OK, are you sure your fan is running right? I am using a different duct: https://www.thingiverse.com/thing:4042492, and had to somewhat increase my fans speeds over what I used with my previous extruder. The place the trouble starts is, of course, just where the object starts to have very short layer times, so the plastic is never getting a chance to cool.
The fan is running fine The problem does not start where the object starts to have very short layer times, the layer times are identical Here is what the finished object looks like, next to another of the failed prints which has a less pronounced version of the issue: Maybe you are thinking that the raised part of the failed print in my previous picture is a higher layer - it is not. The raised part is the same layer as the whole of the rest of the print. At the point I aborted the print, the top of the print should be the same height at all points
You're right... I though that that part sticking out was all alone. Seeing the whole part clarifies the picture. It still looks undercooled, but not (maybe) due to short layer times.
I cannae turn up the cooling any more, it's at 100%, and not only that, it's a 5015 fan now which is bigger than the stock Ender 3 fan, so has more cooling surely than my old extruder which printed fine with the same filament at the same temperature. In fact, the part being printed is the BLTouch mount for the Hemera, which finished printing immediately before I installed the Hemera, so there is literally hours between those prints I will say though that I think the filament has too much moisture in it, I hear pops (Which I don't remember hearing with the old extruder) - it's the first filament I bought, but it's the only PLA that I have, the only other filament I have is TPU (Which is what I normally print in), and that's harder to print, so I thought I would stick with the PLA for test prints I am left wondering though if it was not an issue for the old extruder, why is it an issue for the Hemera?
PS yeah I did print the HangTight duct myself first, but it obscures where I wanted to mount the BLTouch, so I ended up having to revert back to the original extruder and printed a different duct.
PPS - I have a print of it running now, and it has gotten past that point with no significant problems so far. I reflashed the firmware and instead of using BigTreeTech's Marlin fork as a base, I used this one which seems to be using a more recent Marlin build. Dunno if that is what has made the difference or some other factor, but hopefully the problem is gone.
So it completed, but it ain't hugely pretty Strangely, the piece that is sitting on it's own on the left was actually a support for the overhanging part of the print on the left of the image - it disappeared mid-print and was off the print bed behind the printer. One can only assume it got blown away by the part cooling fan? Either that or I have a poltergeist
Or it got picked up by the nozzle, dropped, then blown by the fan. FWIW, if I ever have tiny bits of support like that I tend to stick a small single layer shape underneath them. Judging by the support on the right of the model in the pic, I suspect your support generation is also being overzealous as that's a perfectly OK angle to print with supports.
It's just me being over-zealous with the supports - for the little extra time it takes, better to be safe than sorry in my book It didn't get picked up by the nozzle as it happened after it had started printing layers above it
