bed not level - mesh bed leveling does not compenstate
 

bed not level - mesh bed leveling does not compenstate  

Page 1 / 2
  RSS
mike.m13
(@mike-m13)
Active Member

I can't seem to get a good Z offset value - it looks like there is just too much difference from the left side to the right side of the bed. I have cleaned the steel sheet to make sure its sitting correctly. I thought the mesh bed leveling routine was supposed to compensate for this sort of situation. Here is my print that takes up about 2/3rds of the bed, you can see the left side is way too high, leaving gaps between lines, and the right side is way to low, causing very high adhesion to the bed which makes removing the prints very hard and in my case on the reverse side of the bed, already damaged the PEI layer. 

I was really hoping this would Just Work out of the box, and super frustrating as I am dealing with almost the exact same issue with my MK3S machine (support is sending me a new PINDA and bed, as my magnets started falling out).

Any advice on how to get the MINI to print level would be greatly appreciated!

(Oh, and upgraded to 4.0.1 just before trying the print pictured below, so that didn't solve it!)

Left side

 

Right side:

Posted : 22/12/2019 10:11 pm
PJR
 pjr
(@pjr)
Antient Member Moderator

Just a couple of things you could check:

The u-bolts holding the bearings.  Make sure that the U-bolts have not been over tightened and that there is a small gap between the top of the U-bolts and the bottom of the bed.  Try sliding a sheet of paper in between the U-bolt and the bed.

Also remove the build plate and make sure that all 9 Torx screws are at least finger-tight.  One or two may have become loose.

 

Peter

Please note: I do not have any affiliation with Prusa Research. Any advices given are offered in good faith. It is your responsibility to ensure that by following my advice you do not suffer or cause injury, damage…...
Posted : 23/12/2019 9:38 am
Filip
(@filip)
Member

Double check the XZ-Y axis connection screws - perhaps you overtightened them, which could result in the skew.

Prusa Customer Support...
Posted : 03/01/2020 11:35 am
weak
 weak
(@weak)
Eminent Member

@filip It would really help if PR would provide tightening torque values (Nm) for all screws. 

Posted : 03/01/2020 11:47 am
NilbogAus liked
joerg.e
(@joerg-e)
Eminent Member

To be honest...

I have (more or less), the same problem.

I used bed visualizer (octoprint) to look, how level the bed is. Answer: Horrible. I tried this and that. Nothing helps. Wanted to to the nylon hack, but found out, that the mini is different and it cannot be done in the way like for the MK3(s).

After using nyloc washer under some bad screws its now a bit better (but not flat).

I now expected that Mesh Leveling is compensating and i will have a decent enough print. But after printing a test (with 6 sheets, 1 layer high - https://www.thingiverse.com/thing:3199976) i can see, that some spots (right/left side) of my print are too far away from the nozzle and the other parts a bit to near the nozzle.

Can it be, that the Mesh Leveling in Marlin is not working right? see in addition: https://github.com/MarlinFirmware/Marlin/issues/16397

Here is my mesh:

Recv: Bilinear Leveling Grid:
Recv: 0 1 2 3
Recv: 0 -0.207 -0.078 +0.014 -0.075
Recv: 1 -0.141 -0.066 -0.045 -0.010
Recv: 2 -0.094 -0.174 -0.144 -0.078
Recv: 3 +0.047 -0.179 -0.171 -0.122

Posted : 03/01/2020 1:10 pm
Texy
 texy
(@texy)
Reputable Member
Posted by: @weak

@filip It would really help if PR would provide tightening torque values (Nm) for all screws. 

It would really help if the OP at least responded to advice offered, in order to help others.

 

Posted : 03/01/2020 1:21 pm
joerg.e
(@joerg-e)
Eminent Member
Posted by: @joerg-e

To be honest...

I have (more or less), the same problem.

I used bed visualizer (octoprint) to look, how level the bed is. Answer: Horrible. I tried this and that. Nothing helps. Wanted to to the nylon hack, but found out, that the mini is different and it cannot be done in the way like for the MK3(s).

After using nyloc washer under some bad screws its now a bit better (but not flat).

I now expected that Mesh Leveling is compensating and i will have a decent enough print. But after printing a test (with 6 sheets, 1 layer high - https://www.thingiverse.com/thing:3199976) i can see, that some spots (right/left side) of my print are too far away from the nozzle and the other parts a bit to near the nozzle.

Can it be, that the Mesh Leveling in Marlin is not working right? see in addition: https://github.com/MarlinFirmware/Marlin/issues/16397

Here is my mesh:

Recv: Bilinear Leveling Grid:
Recv: 0 1 2 3
Recv: 0 -0.207 -0.078 +0.014 -0.075
Recv: 1 -0.141 -0.066 -0.045 -0.010
Recv: 2 -0.094 -0.174 -0.144 -0.078
Recv: 3 +0.047 -0.179 -0.171 -0.122

As an addition. My u-bolts are right (checked), my Screws are not overtighten (checked). My Minda seems to work fine. Firmware 4.0.1

This post was modified 10 months ago by joerg.e
Posted : 03/01/2020 1:50 pm
benji
(@benji)
Estimable Member

Perhaps my post here will help: https://forum.prusaprinters.org/forum/general-discussion-announcements-and-releases/avoiding-early-prusa-mini-issues-common-issues/#post-182742

Prusa Mini Silicone Bed Leveling Mod: https://bbbenji.github.io/PMSBLM/...
Posted : 03/01/2020 2:47 pm
pete.b3 liked
joerg.e
(@joerg-e)
Eminent Member
Posted by: @asdf-tgz

Perhaps my post here will help: https://forum.prusaprinters.org/forum/general-discussion-announcements-and-releases/avoiding-early-prusa-mini-issues-common-issues/#post-182742

Thanks for the hint. I have had done something similar with little paper cards. My problem is, that the right side of the bed is too low. If i am printing only the first layer, i can look trough on some parts on the right.

I now tried a variation of the nylock leveling like for the MK3. Its better now. But not really good. My feeling is, that the Auto Bed Leveling isn't working. Otherwise i wouldn't have parts of the bed where the bed is to low (or the nozzle too high). The ABL must compensate for that if the difference isn't too big. Am i wrong?

My Z is leveled as good as possible. But i can see, that the line on the right isn't flat enough, while on the left everything is fine. Its not much, but detectable.

Can someone please try to check  that with e.g.: https://www.thingiverse.com/thing:3199976
(9o% or 95% percent in prusaslicer).

 

Posted : 03/01/2020 9:04 pm
joerg.e
(@joerg-e)
Eminent Member

 

My leveling „nightmare“…

 

My problem is, that some parts of the right side of the bed seems to be too low - the first layer is not ok there (can look trough). In addition, i have the feeling, that ABL is not working because otherwise i shouldn’t have problem because of compensation the ABL is doing with Bilinear(yes, i have a mesh).

 

I am using firmware 4.01

I checked:

1. all screws (bed and others)

2. my u-bolts are ok

3. the cable to minda seems to be ok

4. live Z is set (-.910)

 

I wasn’t able to print using the whole bed without having problems as described

 

I even tried to do a variation of the „nyloc leveling“. I do have better values (see down), but the same problem when i print.

In addition it seems, that i have different values every G29 i initiate

 

This „G28, G29“ commands are done with cold bed and only one after the other without touching the printer in between.

 

1st try:

Recv: Bilinear Leveling Grid:

Recv:       0      1      2      3

Recv:  0 -0.063 -0.014 +0.012 -0.032

Recv:  1 -0.017 -0.002 -0.049 -0.071

Recv:  2 -0.079 -0.061 -0.074 -0.110

Recv:  3 -0.121 +0.012 +0.023 -0.115

 

2nd try:

Recv: Bilinear Leveling Grid:

Recv:       0      1      2      3

Recv:  0 -0.055 +0.032 +0.026 +0.009

Recv:  1 +0.003 +0.016 +0.009 -0.051

Recv:  2 -0.055 -0.032 -0.026 -0.091

Recv:  3 -0.097 +0.068 +0.049 -0.069

 

3rd try:

Recv: Bilinear Leveling Grid:

Recv:       0      1      2      3

Recv:  0 -0.057 +0.017 +0.021 -0.015

Recv:  1 -0.010 +0.004 -0.020 -0.066

Recv:  2 -0.073 -0.050 -0.051 -0.105

Recv:  3 -0.115 +0.043 +0.035 -0.091

 

4th try:

Recv: Bilinear Leveling Grid:

Recv:       0      1      2      3

Recv:  0 -0.061 +0.008 +0.018 -0.020

Recv:  1 -0.013 +0.002 -0.033 -0.073

Recv:  2 -0.068 -0.052 -0.063 -0.113

Recv:  3 -0.109 +0.036 +0.038 -0.100

 

Where can i look into? 

Am i doing something wrong?

At Marlin git i read a lot that other with Marlin 2.x does might have problems with ABL too (Bug?).

I checked with M420 if leveling is activated after G29 - it is.

Yes, in my gcode G28 is followed by a G29. 

 

I would like to use the whole print and didn’t expected such problems (searching since 3 days for a reason).

 

Its not my first printer (i do own Mk3, Cetus modified and an Hephestos with my own compiled Marlin 1.x). 

 

Please help.

 

Joerg

 

Maybe its possible that others test something like that too? (even described earlier)

Posted : 04/01/2020 5:24 pm
mike.m13
(@mike-m13)
Active Member

OP here. u-bolts checked and are not too tight. Also released the MINDA cable from the ziptie, just in case. No change.

I, too, and suspicious of the marlin auto bed leveling. I will run the mesh bed visualizer and report back if it gives inconsistent measurements.

Posted : 06/01/2020 2:16 am
joerg.e
(@joerg-e)
Eminent Member
Posted by: @mike-m13

OP here. u-bolts checked and are not too tight. Also released the MINDA cable from the ziptie, just in case. No change.

I, too, and suspicious of the marlin auto bed leveling. I will run the mesh bed visualizer and report back if it gives inconsistent measurements.

Do you have any news or improvements. I am struggling a lot - even after nyloc hack. ABL is compensating on the right more than on the left. Resulting in to high on right.

Posted : 10/01/2020 8:32 pm
mike.m13
(@mike-m13)
Active Member

@joerg-e

Unfortunately I have had to set the project aside. I am out of time I can donate to the Prusa QA department. Will try again in a month or so. Back to my MK3S for now!

Posted : 10/01/2020 9:05 pm
dtackgrounded-1
(@dtackgrounded-1)
New Member

Have you seen this post on Github regarding Minda probe warm up?

https://github.com/prusa3d/Prusa-Firmware-Buddy/issues/72

Posted : 10/01/2020 9:55 pm
unknwnrider
(@unknwnrider)
Eminent Member

After taking my mini apart to fix a different issue, while reassembling I noticed that if the Z axis is tilted even slightly, it can have a big impact on side to side and front to back height differences. 

the 3 screws that hold the Z/X to the Y axis have to be adjusted properly to make things square. Even then you may have to shim the electronics box to make everything parallel. 

I also found that the hot end cable bundle was too short. When the extruder moved far left it was causing the nozzle to “lift” away from the bed. You can test this by powering off the machine and moving your extruder far left. If it holds on its own, your cable bundle is ok. If it moves back towards the Z axis slightly, your bundle is too short. You can correct this by cutting the 3 zip ties on the Z axis and readjusting the cable bundle. 

hope this helps 

Posted : 11/01/2020 2:02 am
joerg.e
(@joerg-e)
Eminent Member
Posted by: @dtackgrounded-1

Have you seen this post on Github regarding Minda probe warm up?

https://github.com/prusa3d/Prusa-Firmware-Buddy/issues/72

No. Didn't saw that. But very interesting information. Thanks for the hint.

Will check that myself the next day.

Posted : 11/01/2020 7:57 pm
joerg.e
(@joerg-e)
Eminent Member
Posted by: @jeremy-z

After taking my mini apart to fix a different issue, while reassembling I noticed that if the Z axis is tilted even slightly, it can have a big impact on side to side and front to back height differences. 

the 3 screws that hold the Z/X to the Y axis have to be adjusted properly to make things square. Even then you may have to shim the electronics box to make everything parallel. 

I also found that the hot end cable bundle was too short. When the extruder moved far left it was causing the nozzle to “lift” away from the bed. You can test this by powering off the machine and moving your extruder far left. If it holds on its own, your cable bundle is ok. If it moves back towards the Z axis slightly, your bundle is too short. You can correct this by cutting the 3 zip ties on the Z axis and readjusting the cable bundle. 

hope this helps 

Today i bought  an digital spirit level and another tool to check the "squareness". Will check that tomorrow.

I already checked hot end cable etc. These are ok.

Thanks for your input and hints.

Posted : 11/01/2020 8:02 pm
arklan.uthoslin
(@arklan-uthoslin)
Active Member

similar problems here. I swear mesh bed leveling reads the state of the bed, but then when I run a print it doesn't appear to compensate. currently using the silicone bed leveling mod to try to bed flat, but having issues with one screw on the right never reading level, always wanting to be tighter... 

Posted : 14/09/2020 11:01 pm
joan.t
(@joantabb)
Veteran Member Moderator

Hi Arklan, 
Did you sort this by filing the Ubolt? 
or is this still an issue

regards Joan

I try to make safe suggestions,You should understand the context and ensure you are happy that they are safe before attempting to apply my suggestions, what you do, is YOUR responsibility. Location Halifax UK...
Posted : 15/09/2020 11:06 am
arklan.uthoslin
(@arklan-uthoslin)
Active Member

yea, that fixed it.

though it still seems the hotend isn't compensating for the small variance in the bed level the probe picks up.

Posted : 15/09/2020 7:08 pm
Page 1 / 2
Share:

Please Login or Register