Confusing Bridging behaviour - Start in mid-air

[tryptamine 3]

I noticed my bridges always suck, even with the 'best' parameters. 

Upon closer 9inspection, i notices that the first element of the bridge is just printed in mid-air. It is the first thing Cura does in this layer: Start with the INNER part of the bridge, print it in mid-aior, then go outwards, then the normal walls.

Why is this? What can i do about it?

bridgeingBug.curaprofile

[GregValiant 1,492]

Load the model and set Cura up to slice and use the "File | Save Project" command.  Post the 3mf file here.

 

With just the CuraProfile file, and using a model of my own, I can't duplicate the problem.

(You might try changing what looks like "Concentric" bottom to "Lines" and see if it makes a difference.)

[tryptamine 3]

Hey Greg,

thanks for testing. I already use Lines as Bottom pattern as is also expected this to cause a problem.

Here the 3MF:
 

BridgeBug.3mf

[GregValiant 1,492]

I think the problem is that the model has such small "landing zones" for the bridge ends.  When I set Horizontal Expansion to 0.4 everything looks a lot better.

The author of the Calibration Shapes plugin is here.

 

@Cuq if you could take a look at this - it does appear that the inside strings of the bridge come up short of the walls.  In this view the left end appears to be air printing.

Are there other adjustments that can be made besides Horizontal Expansion or scaling the model?

 

 

Here I've cheated and altered the model to give it some more bridge abutment on the inside.

[Cuq 208]

This models is designed to be print with a maximum of  3 Wall Line Count not 5 .   If you use the recommended settings for this model you should not have this issue.

 

https://github.com/5axes/Calibration-Shapes/wiki/Bridge-Spiral-test

 

3 Wall Line Count

 

 

 

Edited January 4, 2023 by Cuq

[GregValiant 1,492]

@Cuq ---- are you telling me that I have to start reading instructions?  I've avoided that like the plague and I'm not likely to start going into my 8th decade.  For one thing, the fonts are too small.

[Cuq 208]

 

[Cuq 208]

13 minutes ago, GregValiant said:

For one thing, the fonts are too small.

 

Ctrl + [Scroll]

[GregValiant 1,492]

Thank you.  Now I see.

That does qualify as another instruction though.  That's 3 in your last 3 posts.

 

I had been using this piece of technology.  I didn't know the scroll/ctrl thingy.

 

 

 

[tryptamine 3]

Thanks so much for the answers.
Still, i just can not figure out bridgeing settings that give me better quality than just brideging disabled.
Here, first an object with bridgeingh settings as in my config file before.

 

Secondly the same object printed without bridgeing, but with a gcode modifier to reduce printing speed to 20 at layer where the bridge is. The latter is much better.

 

3MF of the bridgeing on file also included.

CFFFP_Servohalter.3mf

[tryptamine 3]

Things i noticed:

When using my normal "Inside to outside" Wall ordering, the first lines of bridges are always printed into the air:

 

The more walls i set, the more in air the first green line starts. 


When printing with different wall ordering, things are a lot better.



In my opinion it should be implemented that for bridge layers, wall printing always happens from outside to inside.

I see no scenario where printing into the air as in my screenshot would be good.

[tryptamine 3]

Lastly:

Even with printing from outside to the inside, bridges are still REALLY bad, much worse then without bridgeing enabled.

 

The shell layers arte printed fine. When starting with the top-layers that build the actual bridge, always one one side threre is a retraction BEFORE the nozzle reaches the bridge wall.

This causes one side of the bridge top to be in mid-air, which obviously is really bad:

 

 

To find out, where this is coming from, i first set bridge co3asting to 0, which did not change a thing.

Then i set overall coasting to 0, still the same.

Also my linear advance is disabled at the moment.

What is causing this?

BridgeBug_2.3mf

[Cuq 208]

As your wall need just two lines to be print, set your Wall Line Count to 2 and it should be OK.

 

 

And it will not change the strength of your part.

Edited January 11, 2023 by Cuq

[tryptamine 3]

Setting the walls to 2 for this model indeed helped to print the bridge at it is supposed to be printed.

However, printing with few walls is no option for me, i usually set the wall count to 5 so i have sturdy prints and no infill showing through.

In this case, for every print that has some wall thinner than 4 nozzle widths, briding fails completely.


is this the case for everybody?

Can't i get this specific print to work without having to set the wall count down?

I do not even understand why with 4 walls as in the picture, the retraction happens too early, before the nozzle reaches the yellow wall lines. From the gcode it looks normal:

 

 

 

 

[Cuq 208]

You can use the Skin Overlap Value to change the overlap position between yellow and Green line.

 

[tryptamine 3]

even if i set skin overlap percentage to 80%, i still have the same issue: The damn bottom lines just do not attach to the walls on the right side.
I has retract before skin enabled, disabled this and this also brought no improvement.


I summed up the problems in a more organized way in this post:

As this is apparently a bug in cura and should therefore be made easily accessible to more people.

[Cuq 208]

26 minutes ago, tryptamine said:

As this is apparently a bug in cura and should therefore be made easily accessible to more people.

 

When it's a "Bug" in Cura, Cura Master, SuperSlicer, PrusaSlicer BambuLab Studio and IdeaMaker  You need to call it "Feature" and "Feature Request" 🤣

Edited January 12, 2023 by Cuq