Doubling up on layers bad for cooling

[rpress 1]

I noticed that sometimes Cura doubles up on the layers, I guess this is for speed. For example it will do two layers at once on a isolated feature, before moving on to the next feature.

But this is bad for cooling, and doesn't satisfy the "minimum layer time" setting. When I have a part with some tricky overhangs and things like that it makes me have to run the print slower than I otherwise would need to.

I also noticed that if I'm printing a round part with radial features it will double up the layers on one of the features every time it goes around. There is a noticable reduction in quality on this one feature because it didn't get the same cooling time as the others.

Is there a way to disable this? If not any hints as to where I might find this in the code?

Thanks!

 

[rpress 1]

These two parts printed at the same time, same orientation. The only difference is that the slicer decided to double up the layers on one of them

 

[rpress 1]

This doubling up problem is somewhat unintentional I think, it's just the path optimizer not caring about the layer change.

I've come up with a patch to solve this problem, testing it now.

 

[rpress 1]

My pull request here:

https://github.com/Ultimaker/CuraEngine/pull/150

 

[Daid 306]

This doubling up problem is somewhat unintentional I think, it's just the path optimizer not caring about the layer change.

 

This.

It's pretty much an unintentional side-effect. It's also almost the only dependency that other layers have on each-other in the export gcode step. Making it easier to multi-thread the engine one day.

 

[IRobertI 521]

Can we then assume that this behaviour will not be changed in future versions?

 

[dampmaskin 0]

From my very short experience with the UMO+, I have seen the same. When some layers have a small "island", Cura 14.12 prints the "island" last in one layer, and first in the next layer. This seriously degrades the quality of the "island" part.

I assumed this was an attempt to reduce the number of retractions, but anyway, it does not seem to be good for the print quality.

 

[Daid 306]

Can we then assume that this behaviour will not be changed in future versions?

 

If you mean "I just accepted the patch that changes this behavior"? then yes.

 

[rpress 1]

Thanks Daid!

@IRobertI, I think what Daid is saying is that the patch reduces the interlayer dependence, which will make multi-threading easier in the future, and that's a good thing.

 

[IRobertI 521]

Looks like I misunderstood the post by daid. To me it sounded like the way it was being done now was a good thing for future multi threading and would therefore remain in place.

 

[Daid 306]

Looks like I misunderstood the post by daid. To me it sounded like the way it was being done now was a good thing for future multi threading and would therefore remain in place.

 

No, the fix is good for multi-threading.

Actually, the fix is so good. That I left it in place for the 14.12.1 hotfix release.

EDIT: Download it from http://software.ultimaker.com/

 

[IRobertI 521]

Excellent news. This has bugged me for quite some time, glad to see it's being fixed

 

[aviphysics 9]

Sweet

[Dim3nsioneer 558]

Used it today for a multiple-object-print. Great improvement.