Printer Model: Custom built (ReDuplicator github project)
MCU / Printerboard: BTT SKR mini e3 v3
Host / SBC: Raspberry Pi 5 (2gb)
klippy.log klippy (3).log (3.6 MB)
Describe your issue:
I have my Cartesian/Bedslinger printer and I’m trying to properly get set up. Previously used DuetWifi board with RRF. Mechanically, things seem to be in good shape. Sensorless homing is working quite well. Passing all the checks. Where I’m stuck is trying to get a good first layer squish. I calibrate my z-offset. I’ve done this both via the built in function or just manually. I’ve resorted to manually with a feeler gauge at .3mm lately just to keep my sanity. I come out with about 0.855. That seems pretty reasonable.
If I print a single, 25x25mm square at .25mm layer height, it comes out pretty good looking as far as “squish” goes.
As soon as I do a bed mesh, and try to set my z height to .3mm, I can’t get the feeler gauge in there any more. It is either hitting the bed or is too close to get any feeler gauge I have in there (smallest is .036mm). Hard to tell. So if I run a print, the nozzle is squished way too far into the bed and it’s awful.
If I clear the bed mesh and re-home, everything is back to how I would expect. I set it to .3mm and my feeler gauge is perfect.
I installed KAMP as a friend who runs Klipper had instructed me to since it was a sort of all in one package for line purge and adaptive bed mesh. I’m not sure if my issue is there or not. I did unload the KAMP config and try to run ADAPTIVE bed mesh before a print and had the same issue (assuming that would rule out the KAMP stuff). I’ve tried setting the zero_reference_position, but that doesn’t seem to matter (didn’t expect it to).
It really feels like something about doing BED_MESH_CALIBRATE is throwing my z-offset off or something. Desperate for some help here. Went round and round with ChatGPT and it wasn’t helpful. Posted on Reddit and someone said they thought they saw this was a bug or something?
In particular index 0,5,6: When the Nozzle is above the bed but the probe is not, it’s obvious that the nozzle goes into the bed. For detailed info see here: Understanding X and Y Axes Limits and Homing
@EddyMI3D
Thanks for the response. I’ll re-read the set up as maybe I’ve messed something up.
I’ve included a picture of my setup with the X-axis homed.
If you are looking from the front, when the X-axis hits it’s home, extruder is all the way to the left and the nozzle is at -5mm.
Home on the Y actually puts the nozzle at 0, the bottom of the bed.
If looking at it from the front, my probe is attached to the front of my extruder motor, placing it well off the plate in front, and to the right of the nozzle.
So when I watch it do the mesh, it probes very close to 0,0 on that first index. When I watch it, all the probe points are on the bed. If I home all and tell it to go to 0, 0 it goes to the correct spot on the bed.
Again, I’ll re-read. It’s entirely possible I’ve screwed up. I thought I had it, as it looks like it’s doing what I was expecting.
I think everything is set up correctly after re-reading. Otherwise, I’m just not understanding something?
I did have one interesting interaction that I will try to recreate today.
After a bed mesh, I homed all and told my z to go to .5mm and I couldn’t fit my gauge under it. I can’t remember if I then moved it to .3mm or not just to see, but Fluidd went to the reconnecting screen. When it came back, the z said it was at 0.023mm. Which I could believe. The smallest feeler gauge I had is .036mm and it wouldn’t fit. I moved it up to .3mm and my .3mm feeler gauge felt right. Not sure if that’s helpful information, but yeah. I will do some more testing later today when I can and see if I can recreate this part of the issue.
In the mean time, if there’s a way to get more output logged somewhere, I’d love to know.
