carvedblock

Any ideas from anyone are highly appreciated. I hoped maybe you could chime in @smartavionics if you have a minute? I think you're the de-facto resident forum experts on shells, wall overlaps, and such things. PS: S3D and Slic3r dont't do this and print the model just fine.

Hi, Please help me understand. I make a model with a shell exactly 1.2mm thick. I use .4mm nozzle with .4mm wall. Yet, Cura tries to fit in two inner walls, or at least moves twice in the inner space (not marked as travel move). Even though: - combing is off - and I even turned on outer wall before inner with minimum wall flow it 100% - compensate wall overlap is off What gives? Thanks! This produced ugly blobs on my print and I really can't get rid of them (This is an edge case. I normally have no issue with blobs, even though I use a long and hot meltzone with stringy PETG for part strength.) I am attaching the cura project and gcode and the stl so anyone can play. 3walls.3mf 3walls.gcode shell1.2mm.stl

I have not checked, but this may extend into the Sura Settings Guide Plugin; may be worth checking.

On this page https://ultimaker.com/en/resources/52343-shell The Filter out tiny gaps description is wrong. It says Filter out tiny gaps Some gaps between walls are so small that the extra travel path is not worth the extra extrusion to compensate for it. Enable this feature to fill those minuscule gaps anyway, if needed. But it should be Filter out tiny gaps Some gaps between walls are so small that the extra travel path is not worth the extra extrusion to compensate for it. Disable this feature to fill those minuscule gaps anyway, if needed.

Puzzling times. The dividing with a zero was a bit of a wishful hack, when I was not able to make the IF statement work... Yet. No errors (with any of the mentioned constructs - including the dividing by zero) The only thing log tells me when I switch the temperature is 2018-07-12 07:32:52,134 - DEBUG - [MainThread] cura.Machines.MachineErrorChecker._rescheduleCheck [129]: New error check scheduled. 2018-07-12 07:32:52,584 - INFO - [MainThread] cura.Machines.MachineErrorChecker._setResult [181]: Error check finished, result = False, time = 0.5s --- After embarrassingly many tests, I found something! Many of the constructs (I did not test all) work if used with material_bed_temperature_layer_0, but not when used with material_bed_temperature ! You probably have an idea what's going on here? I don't see anything. Maybe it has to do with some default calculations, but I tried various options and I see no interaction. Here's a limited case that does not work: I have a cfg quality file with these lines this works fine --------------- material_bed_temperature = default_material_bed_temperature material_bed_temperature_layer_0 = =material_bed_temperature + material_bed_temperature / 13 this alwas evaluates to +5, never to 0 (I tried many variations on this theme, this is just one of them) -------------------------------------- material_bed_temperature = default_material_bed_temperature material_bed_temperature_layer_0 = =0 if material_bed_temperature = 0 else material_bed_temperature + 5 It seems almost like Cura has issues with seeing material_bed_temperature as zero? (Because calculations with material_bed_temperature work fine, just the if evaluation seems to be failing with no error). (Well, most. It seems that more complex calculations fail, but that could be an error of syntax.) SO ?Both of the IF statements work (mine and ghostkeepers simplified one) ?Everything seems to work with that (other than the dividing with zero, which was another issue that made the testing pattern harder to see). ?no clue why material_bed_temperature does not work all around

Awesome stuff @ahoeben. That works, and I understand. Python strings all the way! However, something still eludes me. The following examples do not work even though in a regular python shell I see no issues. 1. If statement "'cool' if material_bed_temperature > 0 else 'very cool'" - I actually use a similar construct in a cfg quality profile with no issues. I'm not sure why it would not work here. ¯\_(ツ)_/¯ 2. moar math "'M190 S' + str(floor(45*material_bed_temperature/material_bed_temperature))" - This should evaluate to 0 if the material_bed_temperature is 0 and to 45 if it is anything else. Yet, it's stuck at 45 no matter what. Thanks for helping me dive into this.

All the arcane techniques...

@ghostkeeper I'm very much interested in this. How would you use the 'value' property in place of the 'default_value' ? (in a printer definition file) This does not work (cura crashes on startup) "machine_start_gcode": { "value": "G29 P" + str(int(machine_width/25)) }, and this does not work either (not that surprisingly) "machine_start_gcode": { "default_value": "G29 P" + str(int(machine_width/25)) }, The question is... what would work? Thanks!

Fascinating discussion @ghostkeeper. It sounds like in software theory none of this should matter, but because of the way filament flows and swells, the 'larger' nozzle slightly smears the 0.35mm lines and may blend better. I don't really get it though. Cura wise, is there any difference between setting the nozzle diameter to 0.35 and setting the line width to 0.35? (for nozzle of 0.4mm in reality). Also, what about the the case when I set the line width to 0.45 (nozzle is 0.4)? The slight overextrusion should also force lines to blend more, or not? I am still missing a piece of this.

There's no GUI for this. You'll either have to edit your machine def. json or give a shot to the new Printer Settings Plugin

Thanks for posting all this

Finally, I published a short blog post on this. With a few screenshots and a rationale behind why one may want to do this. A bit more accessible to people who are not used to thinking this way. End of this story. Tadaaa.

Thanks. Very useful @nallath. I can confirm that I can now modify the profiles! Howto modify a .curaprofile Open the .curaprofile as a zip archive modify the files re-compress the files ONE WAY: a bit convoluted way to do this with no 3rd party software rename the profile to `profile.zip` instead of `profile.curaprofile` extract the files edit them re-compress into a new .zip archive rename to .curaprofile again. done ANOTHER WAY: simpler, but you need to install the open source and cross platform B1 Free Archiver Open the .curaprofile archive with B1 (no renaming necessary) Double click on the file you wanna edit it will open in your default text editor make you changes and save the program will ask you if you want to save the changes done. (B1 will re-compress the archive in the background)

I'm on Mac. I have tried Atom, Sublime Text, TextEdit... If I make any change to an exported curaprofile, then the re-import fails. I get "Errno 22, invalid argument" Is there any special encoding or line ending or something that gets messed up? Thanks for any pointers!

I wonder why does zig zag, line, etc support print 1st layer as a grid. Yes, it may improve adhesion and strength, but it also make the support way more difficult to remove. Can it be disabled in Cura 2? Or a workaround? Cheers.

Linked to this from a github issue https://github.com/Ultimaker/Cura/issues/854

Same problem here! Here's my hypothesis: In the folder where cura stores settings saved through the menu, the profile includes all the overrides including material temp and retraction and everything. (checked with text editor) Similarly, when I export the .curaprofile, it has all the overrides including material temp and retraction and everything. As of now, Cura has both material and general slicing presets, but you can only manage and save the general presets through the gui. My guess is that material profiles are made to always override general profile. Which sounds like a potential design fallacy since the only way to even see what's in them is text editor. As of now the included material profiles are just messed up. The PLA one does not even contain material print temperature! It only has: material_bed_temperature and retraction_amount. SO SOLUTION THAT WORKS for now is to manually edit the included material profiles in a text editor or add your own so that the there's no retraction settings! then the general slicing profiles do work as expected! Keep it simple. Like this pla: [general]version = 1type = materialname = jb_PLA[settings]material_print_temperature = 210material_bed_temperature = 60

Thanks! Had the same problem and was scratching my head.