UBL with BLT - strange Measurments
February 24, 2021 10:43AM
Hi There,

im from germany, new to the forum and what you would call a 3dp newbie i guess.

I am testing a bugfix-2.0.x build (cloned yesterday) with the ubl feature on my ender3 with 4.2.7 board and blt connected to probe port.

When doing a normal bilinear leveling, results are good (see screenshot from bedvisualizer: [imgur.com]).

When doing a Mesh with 5*5 (normally i would do 7.7 at least but 5*5 is faster for debugging now), Probing 3 times and EXTRA_PROBING active:

#define MULTIPLE_PROBING 3
#define EXTRA_PROBING 1

This is what i get from UBL Mesh-Creation (G29 P1) when sending a G29 T

Recv: Bed Topography Report:

Recv:

Recv: ( 35,200) (200,200)

Recv: 0 1 2 3 4

Recv: 4 | +3.339 +2.513 +2.986 +3.348 .

Recv: |

Recv: 3 | +0.193 +0.008 +0.148 +3.340 .

Recv: |

Recv: 2 | +0.042 +0.016 +0.010 +1.937 .

Recv: |

Recv: 1 | +0.022 +0.045 +0.068 +2.342 .

Recv: |

Recv: 0 | +1.530 +0.432 +0.984 +3.508 .

Recv: 0 1 2 3 4

Recv: ( 35, 35) (200, 35)



Im using an ultrabase (glass) and the bed is manually leveled before. More that 3mm difference seems wrong to me. I tested with another firmwarebuild where i enabled all the manuel mesh leveling stuff and leveled my printer with the same .1 feeler and also "just" a 3x3 grid cause i was in a hurry where i got:

Recv: 3x3 mesh. Z offset: 0.00000
Recv: Measured points:
Recv: 0 1 2
Recv: 0 -0.02500 -0.00000 -0.02500
Recv: 1 +0.02500 +0.00000 -0.02500
Recv: 2 -0.05000 -0.05000 -0.07500

The manual MESH looks more like what i should see, counting my "does the nozze grab the feeler good or to soft/to hard"-differences in.


Is there anything i did not do right or something that is not mentioned in the official "manual" of marlin/ubl - cant explain to myself why the probe should read an increase like that from center to the corners while the normal bilinear leveling is giving me fine and tunable results.

I also did an M48 to check my Probe (cause its "just" a clone) - the standard deviation of 10 probes (with EXTRA_PROBING enabled) is around 0.05 what is higher than the original expects but should not throw +3.XX values into the mesh.
Sorry, only registered users may post in this forum.

Click here to login