OrcaSlicer icon indicating copy to clipboard operation
OrcaSlicer copied to clipboard

General fixes and improvements to CR-10 SE, Ender-3 V3 KE, Ender-3 S1 Plus, & small fix for Ender-3 V3 SE

Open cochcoder opened this issue 2 years ago • 52 comments

This PR is working towards fixing all the errors/bugs in the profiles that I added to Orca Slicer.

TODO:

  • [x] Fix incorrect line width (mentioned by exploitfate in #4653)
  • [x] Fix incorrect line width for CR-10 SE
  • [x] Fix CR-10 SE G-code error (by @bkonosky)
  • [x] Add changes to CR-10 SE from @bkonosky in #4966
  • [x] Fix incorrect build plate model sizing #4786
  • [x] Fix the incorrect bed size for the Ender-3 S1 Plus 0.8 nozzle type
  • [x] Make build plate model for Ender-3 S1 Plus more accurate
  • [x] Fix low default acceleration in CR-10 SE profile (mentioned here)
  • [x] Fix CR-10 SE build area
  • [x] minor Ender-3 V3 KE improvements (as mentioned by @Orel-A)

Closes #4786

cochcoder avatar Mar 30 '24 18:03 cochcoder

Yep, never rename an existing file or you will break compatibility ;-) That's what happened with one of your previous commits. I know it's tempting to rename files and uniformize names but refrain from doing that. I made the same mistakes ;-)

bistory avatar Mar 31 '24 11:03 bistory

I am having this strange issue where even though I changed the start G-Code and line width to be correct, they don't apply when I build the program, any help with this would be greatly appreciated!

cochcoder avatar Apr 01 '24 23:04 cochcoder

There is a bug when settings 0% on retract amount before wipe. It will treat it like we set 52% amount before wipe. On BambuStudio the bug isn't present.

Orel-A avatar Apr 15 '24 14:04 Orel-A

There is a bug when settings 0% on retract amount before wipe. It will treat it like we set 52% amount before wipe.

On BambuStudio the bug isn't present.

Is this bug also present in the main build or just in this PR?

cochcoder avatar Apr 16 '24 17:04 cochcoder

There is a bug when settings 0% on retract amount before wipe. It will treat it like we set 52% amount before wipe. On BambuStudio the bug isn't present.

Is this bug also present in the main build or just in this PR?

Main build: image

Orel-A avatar Apr 16 '24 19:04 Orel-A

There is a bug when settings 0% on retract amount before wipe. It will treat it like we set 52% amount before wipe. On BambuStudio the bug isn't present.

Is this bug also present in the main build or just in this PR?

Main build: image

So it's present in the main build? If so, it's out of my wheelhouse. Are there any other changes that you think should be made for the Ender-3 V3 KE, or is it all good on that front?

cochcoder avatar Apr 18 '24 00:04 cochcoder

I can't test properly because it's also causing overextrusion and spiderwebs: 20240420_222048 Same filament, same settings on BambuStudio comes clean! So I can't trust my tests with OrcaSlicer.

Lets' revert the changes to X-Y hole compensation and Elephant foot compensation, set them both to zero. Let the end-user determine these values.

Orel-A avatar Apr 20 '24 19:04 Orel-A

So it's present in the main build? If so, it's out of my wheelhouse.

The problem is caused by this pull request: https://github.com/SoftFever/OrcaSlicer/pull/3888

Orel-A avatar Apr 22 '24 07:04 Orel-A

I can't test properly because it's also causing overextrusion and spiderwebs: 20240420_222048 Same filament, same settings on BambuStudio comes clean! So I can't trust my tests with OrcaSlicer.

Lets' revert the changes to X-Y hole compensation and Elephant foot compensation, set them both to zero. Let the end-user determine these values.

Done, let me know if it fixes the issues

cochcoder avatar Apr 22 '24 13:04 cochcoder

We'll heat the nozzle after the G28 Z command since it can take some time if a previous print was tall and it's cause oozing on the middle of the bed.

M220 S100 ;Reset Feedrate
M221 S100 ;Reset Flowrate

M140 S[bed_temperature_initial_layer_single] ;Set bed temp
G28 X Y ;Home XY axes
M190 S[bed_temperature_initial_layer_single] ;Wait for bed temp to stabilize
G28 Z ;Home Z axis & load bed mesh

M104 S[nozzle_temperature_initial_layer] ;Set nozzle temp
G92 E0 ;Reset Extruder
G1 X-2.0 Y20 Z0.3 F5000.0 ;Move to start position
M109 S[nozzle_temperature_initial_layer] ;Wait for nozzle temp to stabilize
G1 Z0.2 ;Lower nozzle to printing height
G1 Y145.0 F1500.0 E15 ;Draw the first line
G1 X-1.7 F5000.0 ;Move to side a little
G1 Y30 F1500.0 E15 ;Draw the second line
G92 E0 ;Reset Extruder

In JSON: "machine_start_gcode": "M220 S100 ;Reset Feedrate\nM221 S100 ;Reset Flowrate\n\nM140 S[bed_temperature_initial_layer_single] ;Set bed temp\nG28 X Y ;Home XY axes\nM190 S[bed_temperature_initial_layer_single] ;Wait for bed temp to stabilize\nG28 Z ;Home Z axis & load bed mesh\n\nM104 S[nozzle_temperature_initial_layer] ;Set nozzle temp\nG92 E0 ;Reset Extruder\nG1 X-2.0 Y20 Z0.3 F5000.0 ;Move to start position\nM109 S[nozzle_temperature_initial_layer] ;Wait for nozzle temp to stabilize\nG1 Z0.2 ;Lower nozzle to printing height\nG1 Y145.0 F1500.0 E15 ;Draw the first line\nG1 X-1.7 F5000.0 ;Move to side a little\nG1 Y30 F1500.0 E15 ;Draw the second line\nG92 E0 ;Reset Extruder"

Regarding the spiderwebs issue, this is a problem with Orca. In BambuStudio after the wiping procedure there is a tiny retraction of around 0.025mm and it's not a bug as some think, that is because during the wiping procedure a thin film of filament is stuck to the nozzle, this small retraction eliminates that.

Anyways, in the meantime, and the small amount of time that I've had for testing; here are some settings that reduce these spiderwebs: We'll set retraction before wipe to 100%: "retract_before_wipe": ["100%"] And we'll disable role based wipe speed: "role_based_wipe_speed": "0" so it will use the 80% speed of the travel speed.

I've also tested 60mm/s retraction speed with zero retract before wipe and a wiping speed of 240mm/s (So that Orca's algorithm will let the whole retraction happen in the wiping procedure) and I couldn't determine which settings are better.

Orel-A avatar Apr 23 '24 21:04 Orel-A

I added the G-Code changes; it also seems that adding "role_based_wipe_speed": "0", causes the build to fail. Can you try without this variable enabled, just in case?

I've also tested 60mm/s retraction speed with zero retract before wipe and a wiping speed of 240mm/s (So that Orca's algorithm will let the whole retraction happen in the wiping procedure) and I couldn't determine which settings are better.

Adding wipe_speed also causes the build to fail, I am open to having a middle ground between 30 and 60 mm/s for retraction speed to prevent a possible scenario of a filament or printer not liking 60mm/s over a long time (I want to keep these settings conservative and more "stock").

cochcoder avatar Apr 25 '24 18:04 cochcoder

You put the variable role_based_wipe_speed in the wrong file. It should be in the process files of the ke. And it's not an array.

Orel-A avatar Apr 25 '24 19:04 Orel-A

Ok, it's in there now. I also upped the retraction speed a little, should the wipe speed also be brought up a little?

cochcoder avatar Apr 25 '24 20:04 cochcoder

0.2mm file?

Orel-A avatar Apr 25 '24 20:04 Orel-A

0.2mm file?

Missed that, sorry

cochcoder avatar Apr 25 '24 20:04 cochcoder

Are there any other changes needed to the Ender-3 V3 KE or is it all good?

cochcoder avatar Apr 25 '24 22:04 cochcoder

Ok, it's in there now. I also upped the retraction speed a little, should the wipe speed also be brought up a little?

For retraction speed; it's recommended PLA 45mm/s, PETG 35mm/s and TPU 25mm/s. So I guess leave it as default at 30mm/s. As for the wipe speed, currently it should be 80% of the travel speed which calculates to 320mm/s. Faster than that and the nozzle will tear off the prints from the bed.

Orel-A avatar Apr 30 '24 09:04 Orel-A

As for the wipe speed, currently it should be 80% of the travel speed which calculates to 320mm/s. Faster than that and the nozzle will tear off the prints from the bed.

I am a little concerned that this might be too much for delicate filaments, what is your experience with this?

cochcoder avatar Apr 30 '24 17:04 cochcoder

I am a little concerned that this might be too much for delicate filaments, what is your experience with this?

My experience is that it could become a problem with over-extrusion and tall prints. For example knocking tree supports.

Orel-A avatar Apr 30 '24 21:04 Orel-A

I am a little concerned that this might be too much for delicate filaments, what is your experience with this?

My experience is that it could become a problem with over-extrusion and tall prints. For example knocking tree supports.

Do you have a speed that wouldn't cause issues with tall prints?

cochcoder avatar Apr 30 '24 21:04 cochcoder

I haven't tested for a definite answer regarding this. However settings the wiping speed to 200mm/s should be safer but over-extrusion will always create issues regardless.

Orel-A avatar Apr 30 '24 22:04 Orel-A

I haven't tested for a definite answer regarding this. However settings the wiping speed to 200mm/s should be safer but over-extrusion will always create issues regardless.

I'll implement this, I just want to make sure that new users won't have any issues when printing tall narrow objects.

cochcoder avatar Apr 30 '24 22:04 cochcoder

If there is nothing else you can think of that's needed. I'll go ahead and mark this PR as ready for review.

cochcoder avatar Apr 30 '24 22:04 cochcoder

you missed the 0.16mm profile.

if there was an option for a button in Orca to click "print & calibrate" it would have been good to add BED_MESH_CALIBRATE after the G28 gcode to have that same effect. same as in Creality Print. would have been good.

Orel-A avatar Apr 30 '24 22:04 Orel-A

Should I try to add BED_MESH_CALIBRATE after the bed temperature reaches temp?

cochcoder avatar Apr 30 '24 22:04 cochcoder

you missed the 0.16mm profile.

Thanks for catching that one!

cochcoder avatar Apr 30 '24 22:04 cochcoder

Should I try to add BED_MESH_CALIBRATE after the bed temperature reaches temp?

The place for it is right after the G28 Z. I'll test tomorrow if BED_MESH_CALIBRATE PROBE_COUNT=6,6 works (the default is 5 by 5 in printer.cfg).

Orel-A avatar Apr 30 '24 22:04 Orel-A

It's working, can you make the changes?

Orel-A avatar May 01 '24 07:05 Orel-A

Please check this build when you can to make sure that everything is working as intended on your side.

cochcoder avatar May 01 '24 11:05 cochcoder

working good, perfect first layer

Orel-A avatar May 01 '24 14:05 Orel-A