Slashee_the_Cow

So, I'm sorry it's not a plugin, just a post-processing script... but on the bright side that means it's just a single Python file so you can mess with it if you want. Although if you need regexes to control your wiping you may want to see a doctor about that.
 
Just put the Python file inside the zip file (can't upload .py files) into Scripts subfolder of your Cura configuration folder, restart Cura and then add it in the post-processing settings.
StopWipingAtZ.zip

I'll second @GregValiant in saying that I get an error while trying to open the file.
You gotta run the initial layer pretty damn slow (I usually have it set to 15mm/s with 50mm/s travel) but unlike TPU you don't need to run the whole thing slowly (I usually run at 45mm/s with 150mm/s travel). Your Z offset needs to be damn near perfect. PETG is some stringy #&%* to deal with, but in a different way to TPU. TPU is pretty much physically impossible to retract sufficiently so almost any travel will leave a string. PETG is stringy in that it always wants to remain a string leading to the nozzle instead of being planted down. Also turn the initial layer flow up to 105-110%. You want to make sure there's enough on the bed for it to stay on the bed. Trying to print anything too fine (layer height/line width) is a fool's game. You're not going to be extruding enough to get it to stick.  
Greg probably has more than me. I've just managed to have more than my fair share of them die under my watch. I refuse to accept responsibility for any of them.
 
"SLOW THE FORK DOWN!!!!" isn't quite as important with PETG as it is with TPU. I'm not sure I've ever needed to slow down to less than 30mm/s (except for one thing which had really intricate details, which I ran at 20mm/s, but almost all of my PETG prints are more about utility than looks), but it never hurts to make sure your acceleration and jerk aren't too high.
 
What temp is that? Sometimes for different things you need different temperatures to get the best of them. Usually I run PETG at 245° with the bed at 80°.
 
I hope you're keeping your filament dry. I'm in Australia so I can beat you for relative humidity (hit 100% plenty of times last summer!). But that's the sort of weather where I leave my filament in the dryer (turned on) while printing - it has holes so you can feed filament out, I use some Bowden tubing to get the filament into the tent (because I don't fancy running unprotected filament through a velcro flap, plus I made a holder that goes on the top bar of the printer to hold several Bowden tubes in place and a few spool holders nearby with more tubing, helps for multicolour printing). But in that weather even if I printed yesterday I'll run it in the dryer for an hour or two before printing.
 
Possible, but if everything else is printing fine I wouldn't think so.
 
So umm... post a valid project file and I can have a look at it more specifically.

Printing over USB is deprecated and no longer officially supported. It's a relic from a time when printers didn't even have enough brains to read files from an SD card.
 
Unless you have a dedicated system (like OctoPrint running on a Raspberry Pi or a basic PC) it's a bad idea anyway since if your computer crashes, boom, failed print. Also if your CPU is being stressed it could take time for it to send commands which will probably end up as blobs while your printer just sits there waiting to be told what to do.

Go to the Cura configuration folder (in Cura go to Help > Show Configuration Folder), then you want to go to:
plugins\CuraEngineTiledInfill\CuraEngineTiledInfill\tiles\
Probably need to restart Cura to get it to pick up the new pattern.
 
In case the configuration folder doesn't show up, for a Mac:
 

In Cura, just go to Help > Show Configuration Folder
If you'd rather do things the manual way, they're in %APPDATA%\cura
There's a folder for each version, first time you run a new version it copies the folder from the most recent previous version.

I'm not sure there's that many you can change other than making sure you have enough walls and skin layers.
Probably wouldn't hurt to turn up Top/Bottom > Skin Overlap (which makes it expand slightly into the walls.
And maybe turn up Material > Flow > Outer Wall Flow a bit (I'd start testing at 110%) to home some will ooze down just a little and help cover up any layer gaps (although if you have enough walls that should close up any potential gap).
 
And as always remember: small scale testing is your friend. Make a shot glass before you try a yard glass.

I think the problem is that the holders don't quite line up with the base, at least once they're rounded to the nearest layer height and line width, as they're separate objects in the STL.
 
If you turn on Mesh Fixes > Remove All Holes then that makes the holders merge with the main part of model:


 
If you set Experimental > Slicing Tolerance to Inclusive then that makes sure all models are at least as big as they are in STL so it treats them as separate objects but it rounds up so that the ring holders definitely fit in the holes:

and rounds up the layers so there's no missing layer (orthographic view):

 
Of the two, I would pick the Remove All Holes option. It'll give you a cleaner looking print.

This isn't really an "Ender-3 Help Forum" (although apparently UltiMaker support get an overwhelming number of requests for help with other manufacturers' printers just because UM are responsible for Cura).
 
It's probably a blockage somewhere in the system, but there's a few places it can be. It might also not be that. With the demise of 2D printers something had to take over as the device which is impossible to diagnose a problem with and 3D printers were the obvious next step.
 
Best case scenario, hardware wise: clogged Bowden tube. Pull the tube out of the extruder and see if there's anything blocking it. This isn't the only thing a blocked tube can look like, especially if you don't use purple filament, but I love any excuse to break out my macro lens:

You can see that some filament might be able to make it through, but it ain't gonna work.
 
Make sure the hardware is cool, then remove the Bowden tube from the hot end. If it looks like there's anything in the tube, cut it back an inch or two behind where the blockage is and reinstall it into the hot end (push it in as far as it'll go) and if it has a retaining clip put it back on.
 
Also, and somewhat worse, hot end is clogged. Fairly simple test, after making sure your Bowden tube has nothing blocking it, just head up the hot end and try to load some filament (something like PLA, not something soft like TPU). If you feel resistance trying to push the filament through the hot end, it's possible there's something clogging the hot end.
 
Fixing that is beyond my knowledge base, sometimes you can remove the Bowden tube, heat up the hot end and have a little poke with something about 1.75mm in diameter and heat resistant, but if you feel much resistance then stop, that's just a slightly fancier version of the "loading the filament" test.
 
If your hot end is clogged, you might be able to try a cold extrusion or you might have to disassemble it. Please seek information from people who know how to do it (i.e. not me) before you try anything like that.

A post processor is the closest you're going to get.
(n.b. the screenshots here are from Windows, but on other platforms the process is the same)
Download and open the zip file In Cura, go to Help > Show Configuration Folder Inside that folder, go into the scripts folder


(I have a whole bunch that don't come with Cura, so don't worry if yours doesn't have many files) Drag the script from the zip file into the scripts folder
Restart Cura Load your model/project Go to Extensions > Post Processing > Modify G-Code
Click Add a script
Find Remove Spiralize Top in the list and click it

(again, I have a bunch in there you won't) It should now appear in the list, so click Close
Slice your model and save the gcode Note: The script will probably remain enabled (whatever scripts you last used are) if you create a new project, and it'll mess up the top layer of anything which isn't spiralised. To remove it, just go to the post processing scripts window and click the X on its line. You can add it again through that window next time you make something spiralised.
 
Edit: Noticed a bug in it. If you downloaded it in the first few minutes this post was up, download the new version.
 
Edit #2 (5 hours after posting): Had only tested it on @Jonysix's .3mf. Turns out it didn't work if you have retractions on. Bit of a hack, but hopefully it now works. Current version is now v3.
 
RemoveSpiralizeTop_v3.zip

Look at the prepare tab: your model is way broken:

The blue and pink dot pattern generally means faces are either missing or have their normals flipped.
 
If you're running Windows, Microsoft's "3D Builder" app (I think it comes with Windows 10 but you need to install it manually in Windows 11) is actually great at fixing models (as well as doing things like simple cuts and a bunch of other things).
 
And because I just made some brownies and I'm in a good mood... here you go:
texas1_fixed.3mf

You mean like this?

This requires slightly altering Cura's interface code. Easy on Windows. Not sure about Linux (you'd have to edit then repack the AppImage). No clue on macOS.
 
Anyway, the file you need to change is:
C:\Program Files\UltiMaker Cura 5.7.0\share\cura\resources\qml\ObjectSelector.qml
Then you're looking for (what is at least for me) line 110:

(just showing the whole thing in context so if it's not line 110 for you, where it is among the whole file)
By default that line is:
height: Math.min(contentHeight, maximumHeight) You just need to change it to:
height: contentHeight  
You might want to bookmark this page, because this will need to be changed with every new version of Cura you install.

In theory that gcode shouldn't heat the bed any slower than just setting the target temperature initially - what it's basically saying is "wait for bed to reach 15°, wait for bed to reach 17°, wait for bed to reach 19°, etc." but it's not pausing when it waits (the difference may be if your printer uses a power curve to warm up the bed instead of just doing it linearly). Using hard-coded numbers like that in your startup gcode isn't a good idea in case you want to print at a different temperature, like the way you have it set up, if you only want the bed to be 60°, it'll warm up to 75° and immediately start printing before it cools - you need to use the R parameter instead of the S parameter if you want to wait for it to cool or heat (S only waits to heat, so since it's already above the target temperature it'll keep going immediately).
 
Unfortunately there isn't any gcode to control the rate at which the bed heats up. And while @jaysenodell means well and yes what they're suggesting can be done, OctoPrint is their solution to everything, and in many cases it's akin to using a sledgehammer to get a nail into wood (n.b. please don't try that).
 
I would try using startup gcode like this:
M190 S{material_bed_temperature_layer_0 * 0.1} ; Wait for bed to get to 10% of target G4 S3 ; Wait 3 seconds M190 S{material_bed_temperature_layer_0 * 0.2} ; Wait for bed to get to 20% of target G4 S3 ; Wait 3 seconds M190 S{material_bed_temperature_layer_0 * 0.3} ; Wait for bed to get to 30% of target G4 S3 ; Wait 3 seconds M190 S{material_bed_temperature_layer_0 * 0.4} ; Wait for bed to get to 40% of target G4 S3 ; Wait 3 seconds M190 S{material_bed_temperature_layer_0 * 0.5} ; Wait for bed to get to 50% of target G4 S3 ; Wait 3 seconds M190 S{material_bed_temperature_layer_0 * 0.6} ; Wait for bed to get to 60% of target G4 S3 ; Wait 3 seconds M190 S{material_bed_temperature_layer_0 * 0.7} ; Wait for bed to get to 70% of target G4 S3 ; Wait 3 seconds M190 S{material_bed_temperature_layer_0 * 0.8} ; Wait for bed to get to 80% of target G4 S3 ; Wait 3 seconds M190 S{material_bed_temperature_layer_0 * 0.9} ; Wait for bed to get to 90% of target G4 S3 ; Wait 3 seconds M190 S{material_bed_temperature_layer_0} ; Wait for bed to get to target Okay that's still a lot of lines. But I get the feeling you're smart enough to figure out how to add/remove steps, change the target percentages or the wait times if you want it a bit more controlled than that. And yeah, those waits will add nearly 30 seconds to the process. But that's better the blowing the UPS in 30 seconds.

You need to redesign your model. Even if it printed strictly outside to inside, the weight of that layer hanging in midair is just going to fall down because there's nothing supporting it.
 
I've done pretty much exactly this sort of thing myself: you need to chamfer above the hole so that each layer can overhang slightly further in to close the gap:

It's because Support > Support X/Y Distance is high enough that the radius of the support is higher than the radius of the part below, but not higher than than the radius of the part below + its support distance, so it won't print below where it is because it would be too close to the part below.
 
The best fix is probably to increase Support > Support Horizontal Expansion to 1.4mm (not a scientific calculation, just played with the number until I got the result I wanted) or higher, this way it will expand so it's large enough to go around the lower part:

This will keep a decent distance around the lower part so you definitely wouldn't have to clean it up:

 
You could also lower Support X/Y Distance (and Minimum Support X/Y Distance) to 0.2mm:

but this will get the support very close to the part below:

 
And about trees: works fine for me, but every Cura installation is its own beast. Try upgrading to 5.7, there have been tree improvements in the last few releases.

Have you tried running an ABL from the printer's control panel to see if it will reset it internally or something? Enders can be finnicky beasts

A 0.1mm line width is incredibly hard to achieve on a 0.4mm nozzle. The general guideline is 60-150% of the nozzle diameter, which means 0.24-0.6mm for a 0.4mm nozzle. That will also make a 0.25mm thick pin impossible because because you can only print in multiples of the line width.
 
You also need to allow for tolerance: you can't print a pin the exact same size as the hole. It will be a literally impossibly tight squeeze. I usually have a tolerance of about 0.3mm to be able to put parts together.
 
Your acceleration rate is too high for precise work (it should probably be about 500mm/s²) and your jerk is far too high (should probably be about 4mm/s instead of 40). If these settings increase the print time significantly, remember: good print > fast print.
 
To have the best chances of printing something like this, you'd need a smaller nozzle. 0.2mm are fairly common. 0.1mm can be found but I've heard they're very hard to work with without just getting clogged.
 
But ultimately, I think the precision you're trying to pull off is just going to be literally impossible to achieve on any consumer grade FDM printer.

Unfortunately while I seem to be the resident TPU weirdo I'm not sure I can offer too much advice since I've never tried anything other than 95A (not impossible to find softer, but not as easy as going to Amazon and ordering some 95A, plus shipping costs in Australia tend to be on the high end so I make plenty of use of my Prime membership).
 
I'd be very much looking at retraction first - specifically, avoiding it wherever possible (and maybe limiting it using Travel > Maximum Retraction Count and Minimum Extrusion Distance Window and/or increasing Travel > Retraction Minimum Travel). 95A is soft enough that it can get ground up if it goes back and forth through the gears too much (hence why I often opt for strings rather than a failed print) so I'm guessing softer is worse.
 
But just looking at layer 11:
 
That has 17 retraction moves (the ones in light blue) in not a huge amount of filament. A lot of them are very close together (in the support interface).
 
Layer 18:

15 retraction moves.
 
Or I could be completely wrong and it's something else entirely. Make sure you're printing at the right temp and slow enough (I usually set it to 20mm/s for everything).

For most of the Cura configuration files there aren't really any guides.
 
In your case you're using the dark theme which inherits from the light theme, which means anything not specified by the dark theme (like sizes) comes from the light theme.
So in the light theme, you need to change line 511:
"print_setup_widget": [38.0, 30.0], I don't know what units those are off the top of my head. But the important part is that if you make the numbers lower, things get smaller.

Not exactly. It might just run straight through and not fill the nozzle chamber, so you might get small underextruded or overextruded bits if the extruder doesn't keep up perfectly, which it probably won't because of how the motors in it have an amount they move per step so it can't do a tiny adjustment up or down.
 
Normally when printing the flow rate is enough that you'll fill the nozzle chamber, which then basically acts as a "buffer" of sorts to make it come out at the intended rate, but printing a very short layer on a 0.8mm nozzle isn't enough flow to fill the chamber, so there's no buffer to keep it even.
 
Another thing is that there's a lot more room in the nozzle than will be taken up by a 0.8mm line at a 0.12mm layer height, so it won't necessarily come out accurately in position.

If you added the printer through the Add Printer wizard in Cura, then the print bed dimensions should be correct and you don't need to do anything.
 
Why, does something not seem right?

Thanks... that's exactly the file I needed.
 
Fixing it is trickier than I thought it might be. The idea with support interface is that by default there's an air gap of one layer below a floor or above a roof (so that it comes off more easily), but it doesn't seem to be taking into account being on the baseplate (which should give it regular support below, except that your height is exactly what it would use for interface.
 
Unsatisfactory solution #1: Disable Support > Enable Support Interface > Enable Support Roof

Gets us the base support we'll use, but it also means we have no interface so supports are going to be harder to remove (especially little ones like this).
 
Unsatisfactory solution #2: Set Support > Support Structure to Normal

That's exactly how the support should look in that situation. It also gives us the problem that there's a bunch of tiny holes that will get filled with support which is going to be a pain to remove which wouldn't be there if we were using trees.
 
Somewhat satisfactory solution #1: Set Support > Support Interface Thickness > Support Roof Thickness to 0.4mm

This makes the support roof short enough that it'll print regular support beneath it to hold it up. Is it going to make a huge difference having two roof layers instead of the default four? In a model like this, probably not. 
 
Effective but really, really fiddly solution #1: Add supports manually

I've only done it for one section in this screenshot, but you can create a support blocker, move and scale it so that it's the right position and size, and use the Per Model Settings tool to set the Mesh type to "Print as support". But you'd need to use several here due to the shapes the support area goes in.
 
And, umm... that's all I can think of for now. I would definitely advise going with the "lower support roof thickness" option.

The problem is in your model:

(enhanced contrast to show the different shading of different faces)
 
The file is also technically invalid, but not bad enough to prevent it from being loaded. That may be the cause of the wavy lines though (as that face is defined as a bunch of columns rather than the simplicity of just using two triangles).
 
What program did you use to create it? SketchUp is sort of considered a running joke about how terrible the files it exports are and Blender (or other mesh-based modelling programs) don't always get it quite right, because it's not what they're designed for. Ideally for structures made of basic shapes like this you would use a CAD program.

It's as designed. Some people run multiple versions of Cura side by side, for various reasons, some people run into bugs in certain versions, sometimes if they have more than one printer, one will only work in an older version of Cura, that sort of thing. Personally I have five versions of Cura installed side by side to help test problems are people are having so I can use the same version as them.

It's local. Just go to Help > Show Configuration Folder, it'll open the folder for that version (each version uses its own folder). Whenever you install a version of Cura you haven't installed before (uninstalling it doesn't delete the settings folder), it will look for settings folders from older versions and copy them over.
 
The "nuclear option" is to completely reset Cura by going up two levels (in Windows it'll just be %APPDATA%\) and rename the Cura folder to... well anything really. Call it cura_old or something. That way next time you launch Cura it'll be from a clean slate but if you have profiles or anything you need to restore just find them in the old folder and copy it into the new one.

The feature request form is here, feel free to submit it.
 
Personally I don't think it would be adopted - most people (even those doing 3D printing) don't think in base 2 so it would be a very niche feature. Not all printers use 0-255 for their fan scale, also: some go from 0 to 1 so you have to give it the percentage as a fraction.
 
Also to the best of my knowledge there's only about four commands in Marlin that use a 0-255 scale and Cura doesn't have support for three of them:
M42 - Set pin state (designed for connecting custom hardware which doesn't natively support Marlin so you can control it manually by setting the pin state, and PWM (pulse width modulation) pins are set on a scale of 0-255) M106 - Set fan speed (as mentioned above, doesn't always go from 0-255) M150 - Set RGB(W) colour (used for controlling the colour of case lights) M256 - LCD brightness (controls the brightness of the printer's screen, some will use caps like 0-100 but typically 0 is the dimmest and 255 is the brightest) It would be possible to use a post-processing script to convert a fan speed set in Cura on a 0-255 scale to an absolute value (you'd just have to enter it as a fraction in Cura because the fan speed is capped at 100%, so your example would be to set it to 0.4 for S4 or 0.138 for S138).
 
Software development rule #8: No matter how hard you think something will be to implement, it will be harder than that to implement. AFAIK Cura currently only has two modes for fan scale: input * 2.55 for most printers or input ÷ 100 for printers that use a 0-1 fan scale. Which one it uses is based on the printer's definition file, so it's attached to the hardware side of things, not a software setting.