#wavy bed mesh

1 messages · Page 1 of 1 (latest)

wispy laurel
#

Getting a wavy bed mesh on my V2.4 StealthChanger
It's an LDO kit, so the bed should be flat. I heat-soaked it at 60°C for 30 mins before measuring. Any ideas what might be causing this artifact?

[bed_mesh] #
mesh_min: 60,95
mesh_max: 300,330
speed: 250
horizontal_move_z: 10
probe_count: 9,9
algorithm: bicubic
##[(7x7)-1] / 2 = 24
##[(5x5)-1] / 2 = 12
##[(9x9)-1] / 2 = 40
#relative_reference_index: 40
zero_reference_position: 150, 150
#relative_reference_index: 15
split_delta_z: 0.0125
move_check_distance: 3
mesh_pps: 4,4
fade_start: 1
fade_end: 10
fade_target: 0
**

1767416723136.jpg Screenshot_20260103_140507_Chrome.jpg
wind pumice
#

I think it might be smart to disable interpolation to see what's actually there, it may be that the bed is actually fairly flat (the overall range of 0.168mm is fine) but it's the interpolation that adds those ripples.

#

The documentation says lagrage can cause oscillation (ripples) and that bicubic is better for those probe counts, but who knows. Try commenting out mesh_pps and algorithm and go again (or if you have a saved mesh profile, just reload klipper after changing settings and it should redo it without interpolation).

#

While it's possible there's some mechanical issue in the gantry causing wavy z motion along the Y axis (when at high X), it does look more like a "math artifact". So the printer is compensating for a z error that isn't there, and causing first layer havoc as a result.

wispy laurel
#

I’ll give it a try. Thanks for the reply!

wind pumice
#

If you're doing bed mesh probing every time, you might want to try a bit fewer probe points and interpolation. But if you're running saved profiles, I think it's better to instead have more probe points and no interpolation. Measurements points are always "more true" than interpolated ones, and with enough probe resolution the regular linear interpolation (which you get no matter what) is good enough. I run mine at 11x11 and no interpolation (for a 300mm bed) and it works great.

#

However, if "weirdness" persists even without interpolation, there's possibly some mechanical issue that needs to be addressed. But since the measurement range is within 0.168mm it's probably all normal, it usually comes out a bit taco/potato-chip shaped. Mostly due to the gantry "sagging" under load and not so much bed being not flat.

willow bramble
willow bramble
wispy laurel
willow bramble
willow bramble