Disable "Ping-Pong" in Cura?

[conz 0]

I don't know, how to call it. Let's say "Ping-Pong" in the direction of the path between slices.

My observation is, that Cura is printing one layer in one direction and than the next layer in the opposite direction.

It is clear, that this behaviour has the adventage of no movement without printing between the layers. But on the other side it has the negative effect, that the new layer will start on the hottest region of the previous layer (the latest printed lines).

Is there an option to deactivate the "Ping-Pong"?

I would like to test the result with a "normal" path, where every layer has the same direction.

That would add an extra move (without printing) from last point on one layer to the start point of the new layer. But the adventage of this behaviour would be, that the next layer would start at the coolest point of the previous layer.

The heat distribution in the material would be a little bit more balanced.

I hope that would give a little bit less bending on steep angles.

 

[gr5 2,173]

I'd like to see something similar. Cura has a "minimum layer time" setting that I typically set to 5 seconds. It should instead never print over a spot that was printed less than 5 seconds ago. This would *also* eliminate Conz's "ping-pong".

If you print something with 4 islands - like you are printing the 4 legs of a chair - it would never print 2 layers in a row (to save time) but instead it would go back to the coldest island.

I understand that the current algorithm saves time but it also causes problems when printing on a layer that is still hot.

 

[david-ganon 2]

can cura print all the model in a circular pattern, like the walls ?

 

[Dim3nsioneer 557]

you may increase the shell thickness to a value which is larger than half of the largest x/y-dimension of your print.

 

[SandervG 1,521]

Isn't this 'comb' you are referring to or maybe I don't understand?

 

[IRobertI 520]

No it's something different Sander. Say you're printing a cylinder (like the speed test, only without using spiralize), it will print one perimeter clockwise and then switch direction for the next layer. So it doubles back immediately onto the bead of plastic it just put down.

And like in gr5's example if you have two islands it will print the first layer on the left island, then move over to the right and print a layer there, it will then put down another layer on the right one before going back to the left where the same thing happens. It does indeed save time as it cuts down on travel movements but on small parts especially it would be better if it jumped back and forth on each layer.

 

[Daid 306]

Outer shells are printed in one direction, internal holes are printed in the other direction (due to how data is stored in the engine)

 

[conz 0]

@Daid, ok, maybe I was not 100% right.

It is more like gr5 has described it. It's the order of the islands. (Not the direction of the print itself).

If you have 4 islands (Numbers 1-4), than the islands actually are in this order:

layer x: 1, 2, 3, 4

layer x+1: 4, 3, 2, 1

layer x+2: 1, 2, 3, 4

...

I would like to test an order more like this:

layer x: 1, 2, 3, 4

layer x+1: 1, 2, 3, 4

 

[primoz1111 0]

 

Hello!

 

i am new to 3D printing i got ma UM2 one week ago and i cant more happy prints ar crazy good.

 

So if i print with no support and in low rez like cure fast default settings except temperature retraction and bed temp twiked i noticed that high overhang is abit problematic due to that default settings i whod like to see x 1,2,3,4 x+1,2,3,4 so that the string in that layer have time to cool down to proper temp so overhand is bether

 

sry for my english

 

RE: Primož

 

 

[eldrick 65]

Say you're printing a cylinder (like the speed test, only without using spiralize), it will print one perimeter clockwise and then switch direction for the next layer. So it doubles back immediately onto the bead of plastic it just put down.

 

I've seen exactly what Robert describes, and thought at the time that it was a bug. It is, AFAIC, and one that negatively affects print quality.

 

[Daid 306]

Ah, yes, this is due to that the Cura tries to minimize the print head movement, and so starts off where it left off at the previous layer.

This will most likely change, where each layer will start at about the same point. (This will also make it possible to multi-thread the engine at some point)

 

[eldrick 65]

It would be strongly preferable if Cura had a Randomize Start Points feature for the outer layer, to avoid leaving visible scars in the print surface where layer starts are vertically aligned, as often happens now. This is particularly obnoxious when printing any cylindrical objects with Cura. The current (and apparently future) implementation gives up significant print quality in favor of trivial print-time reduction.

Take a look at the Kisslicer/Slic3r/Simplify3D implementations - each allows you to randomize starting points, and/or request that start points be restricted to corners or concave areas, and Kisslicer even allows you to specify a range of angles on the print, where starting points can be "hidden" in unobtrusive parts of the piece.

Until at least randomize is available, I cannot use Cura to manufacture either of the sport-fencing products that I manufacture and sell, which is disappointing.

 

[chrii 5]

Hello

I`m having the same problem with cura, since i want to get rid of the very slow skeinforge after years.

Depending on the part the startpoint is "random" or alwas on the opposite side of the layer before.

here are some sample pictures:

The Squirrel has the "random" problem, the leg has the "ping-pong" effekt.

The rest of the surface is great but this spread out starts make the thing not really useable :(

is there a change that there may be a fix or a new option to avoid this will be available soon ?

the rest of Cura is great so i really want to use this software in the future.

Chri

 

[gr5 2,173]

Try printing at half the speed you were printing at previously and see if that makes it go completely away.

The problem is that the printer pauses briefly at a spot before either going to an inner layer or moving Z axis or retracting. If the printer is moving more slowly then the pressure in the head is lower and when it stops, less will leak out (less pressure means less leakage).

Alternatively possibly raise the temperature a bit, but usually lower temperatures improves quality.

 

[chrii 5]

I already tried serval speeds,temperatures and retract and Z are already really really fast, the effect are not blob`s (too much material) they are already too less material because of the speed from retract/Z travel when not printing.

So the effect is already minimized to the minimum, as you can never get the seam complete vanished.

Therefore it would be great if it would possible to change the start/end oder either to a fixed side / a algorythm for finding a corner/ or at least always relatively the same position.

Chri

 

[aviphysics 9]

Hello

I`m having the same problem with cura, since i want to get rid of the very slow skeinforge after years.

Depending on the part the startpoint is "random" or alwas on the opposite side of the layer before.

here are some sample pictures:

The Squirrel has the "random" problem, the leg has the "ping-pong" effekt.

The rest of the surface is great but this spread out starts make the thing not really useable :(

is there a change that there may be a fix or a new option to avoid this will be available soon ?

the rest of Cura is great so i really want to use this software in the future.

Chri

 

On my printer, this can usually be eliminated by printing below 50 mm/s. The problem becomes more pronounced at higher print speeds. The severity is mostly a function of how much oozing occurs during a pause for retraction or for switching between layers.

IIRC if you have any amount of infill, Cura will tend to switch layers when the print head is in the infill, so you won't get oozing on your skin during that transition.

 

[rpress 1]

This problem should be fixed now with my patch.

http://umforum.ultimaker.com/index.php?/topic/8650-doubling-up-on-layers-bad-for-cooling/