Klipper Bed Mesh

[concentric_killa]

Has anyone used this bed mesh module? If so, how do you use it with the print.cfg? Do you heat the bed and then run a series of commands? 

https://github.com/Klipper3d/klipper/blob/master/docs/Bed_Mesh.md

 

 

[Rwhite692]

I followed these instructions:   https://docs.vorondesign.com/tuning/secondary_printer_tuning.html

 

[concentric_killa]

2 hours ago, Rwhite692 said:

I followed these instructions:   https://docs.vorondesign.com/tuning/secondary_printer_tuning.html

Ok, so I managed to add it to my cfg. I was able to run it BED_MESH_CALIBRATE and save it.  How do I visualize the mesh? 

[concentric_killa]

2 hours ago, Rwhite692 said:

I followed these instructions:   https://docs.vorondesign.com/tuning/secondary_printer_tuning.html

NVM, it popped up in Klipper. I didn't see the new tab. 

[ahough]

8 minutes ago, concentric_killa said:

Ok, so I managed to add it to my cfg. I was able to run it BED_MESH_CALIBRATE and save it.  How do I visualize the mesh? 

I  use the heightmap in mainsail. I can't remember what it's called in Fluidd, but that has a similar mapping tool.

[concentric_killa]

16 minutes ago, ahough said:

I  use the heightmap in mainsail. I can't remember what it's called in Fluidd, but that has a similar mapping tool.

 

Thanks, I found it. I didn't realize mainsail had another tab. I added a load bed mesh command to my print start. Should the print head account for this now? or do I need to add some setting in Cura? 

[ahough]

On 2/18/2022 at 4:30 PM, concentric_killa said:

Thanks, I found it. I didn't realize mainsail had another tab. I added a load bed mesh command to my print start. Should the print head account for this now? or do I need to add some setting in Cura? 

If you have that heightmap loaded in Mainsail, it should automatically be accounted for. I didn't add anything to Cura or print_start. Hopefully I'm not doing that wrong.

Update: I did some more research on this. If you run that calibrate from within Mainsail and choose to save the config, it will save the profile with the name "default" but can be renamed if you desire. Klipper will automatically use the profile named "default" if it exists. Otherwise, you'll need to specify which profile you want Klipper to use. So, that's why mine was being taken into account. Hope this helps.

Edited February 20, 2022 by ahough
Adding some information after research

[zogzogzog]

I'm using Klipper and Fluidd on my V2 - I deleted all the bed mesh clear commands in printer.cfg and changed the slicer startup code to specifically load the default mesh.

I'm using a Mandala Rose bed with their kinetic mounts - originally, with a hot bed but before I completed the enclosure, I was seeing flat within .08mm (using the inductive probe).  After enclosing and with a 100C bed and 40+C chamber temp, I see about a .11mm variance across the bed extremes.  That variance pretty much corresponded to my ABS printing experience as there were clearly "sweet areas" on the bed as well as "not so sweet" areas.  With meshing, every place on the bed prints great.

When you mesh, you'll probably see your bed has a "potato chip" sort of shape.  I explained this to the students on the high school robotics team I mentor explaining that pretty much everything in the universe is the shape of a potato chip at some level as it's typically the lowest stress state for "flat" things.  Our printer beds seem to follow that.

I suspect that increase in deviation is from the X extrusion expanding, taking the middle up as it expands, when the entire chamber heats for printing ABS and I plan to add a steel backer to the top side - maybe even tomorrow.  I just plan on using a 400mm carriage rail that I'll trim down and pull the carriage off of.  I'll need to drill and tap hole for securing the cable chain to the backer rail as it won't be able to go direly to the extrusion anymore.  I suspect the bed is still pretty good but the X extrusion is bending up away from the bed rather than the bed bending down away from the extrusion - different cause but same end results.  Doing a backer seems to be a very common mod so I don't think I've discovered anything new here.  After doing that, I'll re-mesh and see where things are at.

Bed meshing is a win in my book

 

Edited February 21, 2022 by zogzogzog

[concentric_killa]

On 2/18/2022 at 2:50 PM, ahough said:

If you have that heightmap loaded in Mainsail, it should automatically be accounted for. I didn't add anything to Cura or print_start. Hopefully I'm not doing that wrong.

Update: I did some more research on this. If you run that calibrate from within Mainsail and choose to save the config, it will save the profile with the name "default" but can be renamed if you desire. Klipper will automatically use the profile named "default" if it exists. Otherwise, you'll need to specify which profile you want Klipper to use. So, that's why mine was being taken into account. Hope this helps.

So I had to re-do the z offset adjustment after doing my mesh but Mainsail is definitely using the mesh. I did a test first layer print and noticed the Z position slightly changing as the hot end traveled its lines. This is a pretty cool feature!

[concentric_killa]

17 hours ago, zogzogzog said:

I'm using Klipper and Fluidd on my V2 - I deleted all the bed mesh clear commands in printer.cfg and changed the slicer startup code to specifically load the default mesh.

I'm using a Mandala Rose bed with their kinetic mounts - originally, with a hot bed but before I completed the enclosure, I was seeing flat within .08mm (using the inductive probe).  After enclosing and with a 100C bed and 40+C chamber temp, I see about a .11mm variance across the bed extremes.  That variance pretty much corresponded to my ABS printing experience as there were clearly "sweet areas" on the bed as well as "not so sweet" areas.  With meshing, every place on the bed prints great.

When you mesh, you'll probably see your bed has a "potato chip" sort of shape.  I explained this to the students on the high school robotics team I mentor explaining that pretty much everything in the universe is the shape of a potato chip at some level as it's typically the lowest stress state for "flat" things.  Our printer beds seem to follow that.

I suspect that increase in deviation is from the X extrusion expanding, taking the middle up as it expands, when the entire chamber heats for printing ABS and I plan to add a steel backer to the top side - maybe even tomorrow.  I just plan on using a 400mm carriage rail that I'll trim down and pull the carriage off of.  I'll need to drill and tap hole for securing the cable chain to the backer rail as it won't be able to go direly to the extrusion anymore.  I suspect the bed is still pretty good but the X extrusion is bending up away from the bed rather than the bed bending down away from the extrusion - different cause but same end results.  Doing a backer seems to be a very common mod so I don't think I've discovered anything new here.  After doing that, I'll re-mesh and see where things are at.

Bed meshing is a win in my book

 

Yea, bed meshing is awesome. I managed to print some things successfully but I noticed the print head would drag on some parts of my prints. I suspected this was due to the bed not being completely flat. 

 

I currently have a mesh for PLA settings. When I start printing ASA/ ABS/ PETG, I will save meshes for those fillaments. Initially I forgot to change the relative reference index, I chose a 7x7 mesh. I forgot to change this to 24.