I'm getting seemingly random lines scattered across the top surface of my prints:

lines1 lines2

Printer: Anycubic i3 mega
Slicer: Cura 3.2.1
Printer chosen in Cura: Prusa i3, Gcode flavor changed to RepRap
Cura Profile: Fine, "Outer before inner walls" enabled

What might be the reason?

  • $\begingroup$ Try turning off "Combing Mode" in Cura. I had the same issue. I turned off combing mode and the issue was gone. I do not understand why. The tool tips seem to point me in the other direction. So my advice can only be to give it a try. $\endgroup$
    – Chaotic
    Mar 19, 2018 at 0:54

4 Answers 4


This is the result of travel moves passing through/over the top surface. The combing setting can prevent this type of move (for walls, infill or top), and z-hop can avoid making any mark during these travels.

Both settings will affect speed. Z-hop will be active on every layer, for example.

  • $\begingroup$ z-hop is used only when transiting locations on a particular layer. If the print is mostly contiguous, as in the images provided, z-hop will be used only when needed. In the case of the photos, the lines represent when it is needed, very few times and with little affect on the speed of the print. $\endgroup$
    – fred_dot_u
    Mar 7, 2018 at 10:22
  • $\begingroup$ I would also subscribe to the notion that apart for uber-complex models, the speed penalty will be negligible. $\endgroup$
    – mac
    Mar 7, 2018 at 11:01
  • $\begingroup$ @mac, for example the models I posted in my last question, where the texture means that no mitigation was needed. I also saw in my research that z-hop causes wear to the machine, but I'm undecided on that. $\endgroup$ Mar 7, 2018 at 11:04

Our local library has a genuine Prusa i3 Mk2.5 that recently had this problem. Because of the number of fingers engaging such a system, it was not immediately discovered that a different profile had been selected in which the Z-hop was turned off.

Z-hop is a feature in which the nozzle lifts slightly (and is height-adjustable) as it moves from one portion of the print to another.

According to my brief research, Cura supports z-hop in the settings. Either it has to be activated or perhaps slightly increased. The aforementioned Prusa works great with 0.5 mm lift.

  • 1
    $\begingroup$ Indeed I did not have z-hop enabled. I will mark the answer accepted after I have tried again with z-hop. $\endgroup$
    – AndreKR
    Mar 7, 2018 at 0:52

Thanks to the answers of fred_dot_u and Sean Houlihane, I did some more research.

There are three features involved here:

  • Combing (follow the already printed path when travelling)
  • Retraction (pull back the filament while travelling)
  • Z-hop (move the nozzle up while travelling)

The relationship between those settings is like this:

  • Retraction is only done when not combing*
  • Z-hop is only done when retracting

Therefore to lift the nozzle when travelling, one must disable combing and enable retraction and z-hop. This prevents the nozzle from scratching the surface.

* it was deemed unnecessary, but that might change in a future version; there is also a plugin that changes this


It seems like the issue may only be that your nozzle has developed a bit of oozing. The g code likes to repeat the same patterns for at least a few layers every time it changes patterns and if it does the same pattern on the top 3-4 layers than by the time it reaches the last layer on the top that pattern will actually be the slightest bit raised above the rest of the layers because of the ooze building up along the same pattern for how many ever layers it followed the same pattern. Your nozzle is than rubbing on that raised pattern when it completes the last layer. This explains the randomness like OP said because the g code may not repeat the same pattern for 3 or more layers at the surface but without reading the gcode it would seemingly be random when and where they occurred. And truthfully if its a small oozing you could reprint this and it could be gone if you changed almost any setting in cura because the g code would be recalculated differently and new patterns would be chosen but so long as the new gcode did not have the same pattern repeat itself on the top 3 or so layers you would not have these raised patterns for the nozzle to rub on when finishing the top last layer.

  • 2
    $\begingroup$ I don't think the problem is what you think it is, but as a side note: it would be great if you could revisit your answer and use paragraphs, commas and other punctuation to make it more readable! :) $\endgroup$
    – mac
    Apr 5, 2018 at 3:16
  • $\begingroup$ Okay @mac since the OP shows no pictures of a successful reprint. I stand by what I said. If he chooses to change any settings and recalculates the g code he may not have these issues appear on this print because his issue is "oozing". And as you so kindly pointed out my answer only has one real grammar mistake and I have fixed that but the 16 other mistakes are advanced issues and I do not think they are of any concern for a reader. I'm not aware of the advanced grammar mistakes in my answer. Feel free to correct it however lets not spam this page anymore. pm is fine $\endgroup$
    – fresh new
    Apr 6, 2018 at 5:06

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .