Differing behavior of 5.4 and 5.3

[Pridanc 4]

I am using CURA 5.4 today so most words stem from 5.4 until noted later on when I tried 5.3

 

Wanting to print some small bits for a project to verify settings etc, I gave it a go.  Using PLA as the simple go-to I pushed print.

Doing this test print, I found the parts curling up and getting knocked off. Hmmm

 

Upon closer inspection, I noticed that the print also looked sketch on top of the curling (warping) so I was batting 1000 here. Not happy as I'd never had PLA warp before I was at a bit of a loss.

 

Not one to casually toss in the towel I created a test matrix (for settings) of individual parameters to see if I could get the print to:

a) stick to the bed and

b) look better. 

 

As time and tests went on (4-5 min per print) I didn't seem to be making any real headway but hadn't reached the end of my test matrix so was not worried.

 

As a relative newbie, I've come to this forum in the past with questions for my education and along the way y'all have taught me to look harder at a slice in preview as there can be all kinds of hidden bits of information in the slice when viewing it up close. So I followed that advice again.

 

Right off the bat when I perused this up-close and personal view from the first print layer, I noticed that the skirt was "stacking" on itself.  I'd never noticed that before so went in to the slicer's Build Plate Adhesion settings for a closer look.  Huh, seems that if the "minimum" length of the skirt is not reached, the slicer just adds layers  to get to the minimum length as set in the slicer.  Reasonable enough IMO.   As best I can tell the length default for the skirt is 250mm but don't quote me on that.  There was no way I needed 250mm of skirt on such a tiny part so I began lowering that min length number yet not seeing expected change each time I re-sliced the part.  For whatever reason, using this 3mf file the setting always left some extra stacked skirt lines in CURA 5.4 no matter what until I raised that min length number to 400mm to which 5.4 added more lines making the skirt wider. I expected more lines wider, but not the strange stacked lines under any circumstance.  Learning!

 

As I continued re-slicing with ever shorter skirt/brim minimum lengths, as I was convinced this was the logical choice, all seemed to work as I thought it would/should at first. Yet the more I messed with this setting the more it was becoming obvious that the slicer in my CURA 5.4  "kind of" followed my requests but then too, did its own thing.

 

It was about this time that I thought "you know, I've never really had print quality issues or have I ever noticed skirt stacking in my previous version of CURA,  5.3" so I decided to open 5.3 and slice my part in it using all default settings from CURA of the Standard Quality .2mm  profile.   Long story short, even with the default 250mm min length, the preview showed only the 3 lines I wanted. No extras, not stacking. For the heck of it I even changed the min number to 0 which is said to tell CURA to ignore this setting which it did leaving me with my requested 3 lines of skirt with no stacking!  And 5.3 did add lines at 400mm min length which I expected but I've always had the correct number of skirt lines in all prints from 5.3 no matter what having never adjusted this min line length before. 

 

After slicing in 5.3, I thn sent this to print on my Ender 3v2 and dang if the print wasn't almost perfect in every way.  

 

Now we get to the question or perhaps it is just me letting y'all know something is odd in 5.4 on my computer.  ( IMO of course)

 

I've included the 3mf file as well as  jpeg snippets of the skirt found when reviewing the same sliced file in CURA 5.3 and then CURA 5.4. The pictures clearly show skirt differences even though the file is the same between them.

 

For grins and giggles, I will be printing this 3mf file (project) from the slice file out of 5.4 to see how it prints vs how it printed out of 5.3.   No question that slicing the project in 5.4 changes the skirt as can be seen, but what else might be different? The visually results might show me that there are other differences.  I'm also going to compare the html "project files" from both slicers to see if there are differences that show up.

 

Many thanks as always and I look forward to hearing back when time permits.  Stay well, PDC

Plate-to-Plate 5.3.3mf

[Slashee_the_Cow 442]

By this point it's very well known that Cura 5.4 has some bugs, especially when it comes to slicing. Something you can do try and mitigate the problem is go to Mesh Fixes > Maximum Resolution. 0.5mm is a good starting point, if that doesn't work try going down a bit at a time.

[GregValiant 1,359]

The "Skirt Stacking" is a new setting in 5.4.0.   I've never had a problem removing the skirt from the bed so I don't need that setting at all.  The setting is in the Bed Adhesion section, it's called "Skirt Height" and the default is "3" layers high.

 

Set it to "1" and you will be back to having the skirts as you expect.

[LePaul 331]

Adding addition layers of Skirt Height in 5.4 is the most moronic change I've seen with Cura.  And I've been here since 2014!  

[Gero 54]

Setting the skirt height back to 1 for each new print is no longer necessary when switching to cura 5.3 or 4.13 for the time being 🙂
but it's standard settings like this, which i have to adjust with every new print, that make using cura unnecessarily inconvenient in the meantime. 

[GregValiant 1,359]

I have a couple of overrides in my definition file.  One more didn't take but a second.

"skirt_height": { "default_value": 1 }

[Slashee_the_Cow 442]

Wait, am I the only one here who saves their own custom quality presets?

[GregValiant 1,359]

That doesn't work for me.  I end up changing them for every print anyway.

I have "Slow PLA", "Fast PLA", and "PETG" and I have a separate Ender 3 Pro installed for TPU because it needs it's own startup gcode.

So every print is a custom setup.