Cura Script to Automatically Probe Only Printed Area

[Baenwort]

@zapta said in Cura Script to Automatically Probe Only Printed Area:

@tcj, thanks for the example and for your patient.

I made a fix and am now getting the output below. The problem existed in the original code posted in this thread, parsing only non negative x and y values.

https://github.com/zapta/misc/blob/master/duet3d_automesh/duet3d_automesh.iml

Does this bring it closer to working for Delta's? Or at least probing a square area of the circular bed?

[zapta]

@Baenwort, what are the special requirements for delta compared to a cartesian?

The current script finds the bounding rectangle of the first layer's print (negative numbers are ok) and generates a M557 command with X and Y ranges and number of probe points.

In case of a delta, the bounding rectangle can have points outside the round print area but I presume the duet will ignore them.

Edit: for delta, if objects are placed in the center, it's possible to compute the radius from (0, 0) and issue a M557 with a R param.

[Red Sand Robot]

I'm also having an issue when getting the post processing script to run. I am getting this:

Post-processing script C:\Users\Jack\AppData\Local\Programs\Python\Python38-32\python.exe "C:\Program Files\PrusaSlicer-2.2.0+win64-202003211145\PrusaSlicer-2.2.0+win64-202003211145\duet3d_automesh.py" on file C:\Users\Jack\Documents\3D Printing\Stuff\Flex Tube Support.gcode failed.
Error code: 1

Any ideas on what I may be doing wrong?
This is my full start gcode script:

M82 ;absolute extrusion mode
G21 ;metric values
G90 ;absolute positioning
M82 ; set extruder to absolute mode
M107 ;start with the fan off
G29 S2; clears existing bed heightmap

; For automesh
M557 TBD  ; parameters will be set automatically
G28; home all

G32; motorized gantry plane correction
G32;
G32; Repeated 3 times to ensure its correct

G29  Mesh Bed compensation 

G1 Z20 F6000 ;move the platform down 15mm
G1 F100 E9 ;extrude 9mm of feed stock
G92 E0 ;zero the extruded length

[OwenD]

@Red-Sand-Robot
Double check the content of the script.
I just tried yesterday and was getting the same error.
The first time I downloaded, I right clicked the link and selected "download link" or some such.
Problem was it downloaded the whole HTML page, not just the python script. (DOH!)
After copying and pasting just the script content it worked fine.

[zapta]

@Red-Sand-Robot, I would suggest to first run it manually from the command line. It will give you more information on what's going on. (after verifying that the script download is ok per OwenD's suggestion)

[Red Sand Robot]

@zapta Command line? As in my computer or PrusaSlicer? Sorry, the programming side of things isn't my forte.

[zapta]

@Red-Sand-Robot, it seems that you work on windows. I meant commands types in the DOS window. I don't have here a Windows computer but others here may be able to explain it better.

https://www.lifewire.com/how-to-open-command-prompt-2618089

[Red Sand Robot]

@zapta Yes, I work on windows. Haven't made the jump to attempt to learn Linux or associated stuff yet though I suppose I should if I ever want to use a RPI for anything useful.

When I ran the script from the command prompt window, I received this output:

C:\Users\Jack>AppData\Local\Programs\Python\Python38-32\python.exe "C:\Users\Jack\Documents\3D Printing\PrusaSlicerScript\duet3d_automesh.py"
usage: duet3d_automesh.py [-h] [--meshable MESHABLE] [--margin MARGIN] [--spacing SPACING]
                          [--min_points {2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49}]
                          [--max_points {2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49}]
                          [--first_layer_start FIRST_LAYER_START] [--first_layer_end FIRST_LAYER_END]
                          file_path
duet3d_automesh.py: error: the following arguments are required: file_path

Does that look like it is functioning correctly? I'm guessing that the file_path argument is talking about the .gcode file, but if not that would indicate an error with how I'm calling the script from what I understand.

[Red Sand Robot]

@zapta Figured out that my main issue was that I never changed the meshable area parameters to match my machine. I guess I just assumed when it says "default" that the script would also parse that stuff from the slicer, but in hindsight the script doesn't function that way, it only looks at the .gcode file after it has been output by the slicer.

Leaving output up as an example of what it looks like the user does not properly define the meshable area in regards to their machine. My machine has -100:100,-100:100 for a meshable area.

C:\Users\Jack>C:\Users\Jack\AppData\Local\Programs\Python\Python38-32\python.exe C:\Users\Jack\Documents\3D-Printing\PrusaSlicerScript\duet3d_automesh.py C:\Users\Jack\Documents\3D-Printing\Stuff\flex_tube_support2.gcode
MESHABLE area: 30.0:280.0,30.0:280.0
Opening gcode file: C:\Users\Jack\Documents\3D-Printing\Stuff\flex_tube_support2.gcode
Read 33095 lines
Parsing state = ParsingState.WAITING_FOR_LAYER1
; Automesh: begin layer 0
Parsing state = ParsingState.IN_LAYER1
; Automesh: begin layer 1
Parsing state = ParsingState.LAYER1_DONE
First layer print areas: -13:13,-48:48
printArea: -13:13,-48:48
Traceback (most recent call last):
  File "C:\Users\Jack\Documents\3D-Printing\PrusaSlicerScript\duet3d_automesh.py", line 371, in <module>
    main()
  File "C:\Users\Jack\Documents\3D-Printing\PrusaSlicerScript\duet3d_automesh.py", line 218, in main
    mesh_area.clip_to(MESHABLE_AREA)
  File "C:\Users\Jack\Documents\3D-Printing\PrusaSlicerScript\duet3d_automesh.py", line 165, in clip_to
    self.x_span.clip_to(other_rect.x_span)
  File "C:\Users\Jack\Documents\3D-Printing\PrusaSlicerScript\duet3d_automesh.py", line 126, in clip_to
    self.__check()
  File "C:\Users\Jack\Documents\3D-Printing\PrusaSlicerScript\duet3d_automesh.py", line 105, in __check
    fatal_error(f'Invalid range value: {self}')
  File "C:\Users\Jack\Documents\3D-Printing\PrusaSlicerScript\duet3d_automesh.py", line 83, in fatal_error
    raise Exception('Fatal error: ' + message)
Exception: Fatal error: Invalid range value: 30.0:23

[zapta]

@Red-Sand-Robot, I think we are almost there

Can you post here your file flex_tube_support2.gcode ? I will try to run it myself and check that error.

Edit: to set your own meshable area, add to the command line the following flag:

--meshable=-100:100,-100:100

[Red Sand Robot]

@zapta Would --meshable=-100:100,-100:100 go after the output path specification in PrusaSlicer? I just manually changed the default in the script.

parser.add_argument('--meshable',
                    default="-100:100,-100:100",
                    help='Bed meshable area x1:x2,y1:y2')

Here is the .gcode file. I have my 0,0 point in the center of the build plate. I know that you previously added support for negative positional values so I don't think that would be the issue
flex_tube_support2.gcode

[tcj]

@Baenwort said in Cura Script to Automatically Probe Only Printed Area:

Does this bring it closer to working for Delta's? Or at least probing a square area of the circular bed?

Yes, this makes it usable for delta's, because
"For Cartesian printers, specify minimum and maximum X and Y values to probe and the probing interval. For Delta printers, specify the probing radius. If you define both, the probing area will be the intersection of the rectangular area and the circle. "
https://duet3d.dozuki.com/Wiki/Gcode#Section_M557_Set_Z_probe_point_or_define_probing_grid

[zapta]

@Red-Sand-Robot, the change that you made in the program, changing the default meshable area to -100:100 is correct. Just remember to apply it again if you will download a newer version of this script.

I run your gcode file and got this which looks ok to me

MESHABLE area: -100.0:100.0,-100.0:100.0
Opening gcode file: flex.gcode
Read 33099 lines
Parsing state = ParsingState.WAITING_FOR_LAYER1
; Automesh: begin layer 0
Parsing state = ParsingState.IN_LAYER1
; Automesh: begin layer 1
Parsing state = ParsingState.LAYER1_DONE
First layer print areas: -13:13,-48:48
printArea: -13:13,-48:48
Will use 3 x 4 mesh points
Marker: M557 X-23:23 Y-58:58 P3:4
Inserted: M557 X-23:23 Y-58:58 P3:4

Does it work for you now or are there still problems?

Edit: for your question the -- flags should be specified immediately after the .py script name.

[Baenwort]

@tcj Great! I'll give it a try on my next print.

Do you think it will use the Delta probe interval for the new area or one specified by the gcode?

[Baenwort]

@tcj So I'm still learning Cura after switching from using MatterControl since I got my Rostock v3. I tried adding the duet3d_automesh.py from zapta's github to the scripts area of the postprocessing folder of Cura's plugin folder.

I do have Python 3 installed (3.8.3 to be exact) but this seems to make no difference to it showing up in Cura.

[zapta]

@Baenwort, can you provide a little bit more info. E.g. what computer you are using (windows? mac? linux?) and how you set up the script in cura.

[Baenwort]

@zapta

Window's but I also tried it with a Ubuntu 20.04 Cura install. Both are Cura 4.6 and the OP and most post processing scripts direct placing the .py file in the plugin post processing directory.

I can get you the exact tree when I get.back to my home tonight.

[sandoz]

I have installed this into Cura 4.62 and using on my Railcore II. When it goes to run it always defines my X end point as being 1800.

M557 X103.850:1800.000 Y115.304:164.705 S10.000 ; Leveling mesh defined by LevelingMeshOptimizer

Railcore of course doesn't like this and then just skips the mesh:

Error: M557: bad grid definition: Too many grid points; suggest increase spacing to 44.2mm
Error: G29: No valid grid defined for bed probing

Is there something I am doing wrong?

[zapta]

@sandoz said in Cura Script to Automatically Probe Only Printed Area:

Is there something I am doing wrong?

My guess is that there is a X1800 somewhere in the gcode that is picked by the script. Can you post the gcode file?

[sandoz]

@zapta Alright.. using your comments as a place to start I found that having my Jerk control turned on in Cura led to some M566 x1800 commands that it was picking up. I turned off Curas control and just set it in the machine settings and now it slices properly.