@joncruz @noiob @Siff I get the feeling it's a little bit of everything and the texture of the PEI plate around the Perimeter looking good was somewhat of a red hering. Trying the things recommended on their own didn't really solve the issue (very slow speed <15mm/s 'seemed' to do something).
Lowering the nozzle by additional 0.05mm, going 10°C+ higher in temps for the first layer and wiping the plate with isoprop once again (or using the very edge that got never used) seem to improve things.
@noiob I do have a probe, the bed gets measured for every print. The cooling is disabled for the first two layers.
@Natanox then I'm confused why it's going off to the side like that, is your nozzle flowing freely?
@adorfer @noiob Already did yesterday, it's behaving the same everywhere. 🫠
I've managed to make it stick by slightly lowering the nozzle further (almost sure it's slightly too low now) and raising the hotend temperature for the first layer by 20°C to 240°C which is above material spec. It still visibly *really wants* to not to though, and I can see the plate structure imprint on the bottom layer being less strong in the middle of the object. No clue why PLA behaves like this.
@Natanox oh, and it kinda looks like the first layer is getting blown away by part cooling, maybe turning that off in the first layer would help (though it'd only be a stopgap vs better bed adhesion)