#Can't get print to adhere to bed.-

30 messages · Page 1 of 1 (latest)

viscid kelp
#

Hardware Setup

Printer: Ender 5 Plus (350x350mm stock bed) (4-5 years old)
Control Board: BIGTREETECH SKR V1.4 Turbo
**Display: **BIGTREETECH TFT35 V3.0 (no longer works with Klipper)
**Stepper Drivers: **TMC2209 (X/Y/E in UART mode, Z in standalone mode due to communication issues)
Auto Bed Leveling: BLTouch probe
Extruder: Stock Bowden setup with Capricorn tubing upgrade
**Z-Axis: **Dual Z motors
**Host: **Raspberry Pi 3 running MainsailOS + Klipper
**Slicer: **Orca Slicer
Filament: PLA

I originally had my printer running Marlin firmware with my BTT Board and Touch Screen, I was having so many issues getting prints to stick/first layer issues after it had sat for some time in my basement unused. I Set Z offset many times did all kinds of things (troubleshooted with the help of AI) and could never get it to print reliably. I thought my 4x4 bed mesh was the issue on my large build plate but didn't want to mess with re-flashing Marlin again.

I decided to switch to Klipper since I can edit the printer config without re-flashing. I installed mainsail os on my RPI3, got the firmware to the printer installed on the board, and got Mainsail/Klipper working connection wise.

**Symptoms **
-Purge line always sticks perfectly to front left of build plate
-Actual print (even a tiny cube) in the center is always 0.3–0.5 mm too high → lines just sit on top and wipe off with a finger
-With BED_MESH_CLEAR the first layer stuck in the center BARELY but eventually failed (but obviously not compensated)
-With mesh loaded → center floats probably 1mm to 2mm to high (this is a guess by looks)

#

**What I have already tried (everything): **
-Fixed y_offset from -8 → 0
-Multiple hot PROBE_CALIBRATE + paper tests (Z=0 is perfect light drag in center)
-Deleted entire SAVE_CONFIG block, forced many fresh hot 8×8 meshes
-Changed mesh limits to 20-330, 30-320, 30-309 (no out-of-range errors anymore)
-150 % first-layer line width (0.60 mm)
-First-layer speed down to 10–15 mm/s
-Bed temp 60 → 70 → 75 °C
-Z-hop completely disabled everywhere
-Retraction on layer change off
-Relative extrusion on, firmware retraction on (tried different combos to)
-Removed every possible absolute Z move after mesh load
-Tried printing with mesh off (sticks but BARELY)
-Tried printing with mesh on (center floats)
-Tried printing without purge line (same result)
-Tried printing with purge line at exact 0.20 mm (purge sticks, print still floats)
-Tried a brand-new process profile in OrcaSlicer
-Tried a brand-new printer profile in OrcaSlicer
-Briefly tried independent dual Z on E1 (reverted because of UART/pin conflicts)
-Cleaned bed with dish soap + IPA multiple times

**Current state **
-Z offset is proven correct (purge + no-mesh prints stick)
-Mesh is fresh, hot, limits correct, no out-of-range errors
-Start G-code is minimal and never overrides Z after mesh load
-First layer line width 150 %, speed 15 mm/s, bed 70–75 °C
-No Z-hop, no slicer offsets, no retraction lifts

Result:
purge line sticks like glue, actual print in the center is still too high and wipes off with a finger. I am now 100% convinced the PEI coating in the center of the glass plate is completely dead/burned off. The edges still have some grip left maybe? Looking for confirmation or any last-ditch ideas before I peel this plate off and slap a new textured PEI sheet on. Or maybe I just take it out back like "Office Space".

Thanks for any help — this has been a nightmare.

woeful flicker
#

Did you add the g29 and m420 s1 command to the startscript? Ohne and because it happens quite a lot make sure that the mesh creation makro and the use mesh makro use the same save slot and dont waste 3 days on a qidi xplus 3 ...

viscid kelp
#

Here is what my start G-Code-

M140 S[first_layer_bed_temperature]
M190 S[first_layer_bed_temperature]
M104 S[first_layer_temperature]
M109 S[first_layer_temperature]
G28
BED_MESH_PROFILE LOAD=default
G92 E0

G0 X10 Y10 F12000
G0 Z0.20 F600 ; drop to exact first-layer height
G1 X10 Y140 E20 F800 ; fat purge line
G92 E0

woeful flicker
#

What you can do is run bed_mesh_ calibrate and hit the save button a couple of times to make sure there is a mesh saved to use and see if it gets better

viscid kelp
#

There is def a mesh saved. It loads in with BED_MESH_PROFILE LOAD=default in my start gcode and I can see the mesh in my heatmap in mainsail. I DID JUST NOTICE that Elephants Foot Compensation WAS enabled in OrcaSlicer..... Im not home but when I get there for lunch im turning that shit off and seeing if that was the issue.. Its just so weird because my purge line goes down very well and sticks.. but as soon as it start the actual print its 0.2-0.4 mm off by the looks of it.. It just isnt squishing it down to the bed.. If i check Z0 in the center of the bed it is perfect. A peice of paper is snug.. Im about to have a free printer for sale lol

#

Appreciate your help so far with this! Its nice to bounce ideas of real people instaed of AI or researching Reddit/forums

#

If its isnt reactivating it.... is there gcode i need to set in my start to re-enable it?

woeful flicker
#

My second guess would be that default position is slot 0 or 1 and mesh calibrate saves to the opposite one

woeful flicker
#

Ooohhhh

#

You have z fade on

#

That could be the issue

viscid kelp
#

Ill try turning that off as well. I didnt think it mattered for the first layer though as it didnt fade until later.. Worth a shot turning off tho!

woeful flicker
#

i hate fade nearly more then creality

#

it seems to never work

slim bough
#

You havent said what bed you have or how you clean it?

viscid kelp
#

@slim bough ive used dawn/water rag and have used ISO rag. Tried hairspray/gluesticks to. It's the stock glass bed. I just ordered a new magnetic bed last night.

@woeful flicker I just turned off Elephant Foot Comp and fade, got a print to stick! It printed very weak/thin though so I think I have a flow rate issue. I tested my extrussion and my 100mm extrusion test pushed out 98. So its close but could be adjusted. Im wondering in Orca somewhere its choking the flow rate down. I JUST moved over to orca from Simplify3D, which is what i used for 8+ years previously. Im getting closer and making progress! Ive spent entirely to long on this so far lol 13+ hours of troubleshooting/testing. Im back at work now so will have to check slicer for flow rate when i get home. But the print itself looks like dogshit. Almost see through walls/layers

woeful flicker
#

your first layer height is .4 mm with is 25% over the max of a .4 nozzle. rule of thumb is always only 3/4 of nozzle diameter because you want squish

#

!level

daring dockBOT
woeful flicker
#

at least from the last gcode i checked

viscid kelp
woeful flicker
#

in your example gcode yesterday the gcode viewer said .4 mm

viscid kelp
#

That wasnt me im not sure where that came from! But anyways I have figured it out!!! In my printer.cnf i had [stepper_z] rotation_distance: 4 that was NOT correct it needed to be 8.. I have no idea how that got messed up and missed for this long but once i finally got a print to stick i noticed it was printing extra extra tall... and that made me think that something was wrong with the x axis lead screw info and sure enough.. that was it.