4
$\begingroup$

I'm a 2-week newbie at 3D printing, working on a new Qidi Xpro machine (that is solid and one that I like). So, I do not want to believe that this issue is caused by my printer itself. I'm hoping that my settings have something to do with it.

The problem is all the filament lines (travel lines, I think) that start at a sharp corner and go somewhere else. Most often, the lines go to another sharp corner but sometimes can intersect the middle of a side. See the image below for many examples. I have drawn white lines parallel to the unwanted filament lines in case they are hard to spot.

The unwanted lines also appear in the infill underneath the surface lines. They look like porous infill grid lines since they should probably not be there, and they do not get a full load of filament extrusion. You could easily say "Oh, they are the result of a "leaky" nozzle with poor retraction, but I think it's more complicated than that. I have done a full load of retraction tests and calibrations to optimize retraction lengths and speeds to minimize hairs.

Here is an image showing the problem:

Unwanted filament trails and infill lines parallel to white lines

An image showing the trails in the infill

My Cura settings seem reasonable for PLA: bed 50C, nozzle 200C, print speed 40mm, travel speed 90mm (100-110 makes no difference), retraction length 8.5-9.5mm (makes no big difference), retraction speed 35mm, infill: density 20% (line, grid, makes no difference), z-seam set to random, retract at new layer = enabled.

I want to believe that something in my settings is telling the machine to extrude a 1/2 or 1/3 amount of filament when it starts those travels from sharp corners to somewhere else. But, I have not been able to solve the problem.

Everything else works fine in a print (IMHO) except for another problem that I described here. I thought I fixed that one, but I saw it occur on the image shown above (inside the infill, under the surface that is shown above).

Does anyone have any ideas that I might try to solve the problem? Thank you.

UPDATE: Here's an image with Zhop when retracted enabled at zhop height 1mm. This is an infill picture, so it cannot be directly compared to the surfaced original. But, the unwanted trails are all still there (although very thin). Maybe 2mm hop height will do it. I will run another test.

With Z-Hop When Retracted = enabled, hop height 1mm

UPDATE:

I ran a second test with zhop 2.0mm, but without success. The problem was still there. (Retraction on, 8.5mm, min distance 0.8mm). Here is a comparison picture. I think the gcode is definitely telling my printer to do what it does, because the problem is not just a random leak. I think I'll try a different slicer in hopes that it generates different gcode.

Zhop comparision, 1.0mm on the right.

UPDATE ON COMBING

Oscar pointed me to the Cura "combing" setting in his comment below. He nailed teh problem perfectly. Combing means "don't retract, and ooze as you please while you move in a straight line to the destination." That's exactly what I've been showing in my images.

The default for Cura combing is enabled. As a newbie, I didn't know enough to turn it off. Worse yet, if combing is enabled ("don't retract"), it prevents "Zhop When Retracted" from zhopping. So, all my zhop experiments did exactly nothing and had zero effect because no zhopping was occurring. After I disabled combing, then for the first time in my life I actually saw what a zhop looked like. (And thus I can assert that zhopping was not happening with combining=enabled.)

Here is an image of my new "perfect" prints thanks to Oscar's pointer on combing. The left two prints have combing off, zhop off. They are almost perfect, inside and out. The infill walls are generally solid and smooth (20% infill) and without the globs and gaps in the image on the right (with combing on).

enter image description here

As a closing note, I think the combing setting was probably also responsible for the problem of missing infill grids in the corner, as described in this question. The problem doesn't happen with combing off. My theory is that with combing on (as it was for that question), the oozing pulls filament out of the nozzle during the travel, so there's not enough filament at the destination to bite and bond to the corner. So, turning combing off solved 3 problems for me: the awful surface lines, the unwanted trails in the infill, and the missing infills in the corner. It also greatly improved the quality of the infill walls, too.

$\endgroup$
  • $\begingroup$ Could it be, that you are overextruding? Maybe you can also try a print with lower temperatures such has 190 and 180. Maybe your thermristor is off.. $\endgroup$ – dgrat Sep 26 '18 at 10:10
  • $\begingroup$ @dgrat, thank you for your help. I did try reducing print temp to 195, but it made no difference. It could be that my thermistor is off and that my filament is runny. (But, the machine is <2 wks old...) If the thermistor was off, then everything would be runny, yes? But for the most part, my prints and walls are smooth without overextrusion issues. I printed one of those calibration prints from Thingiverse, and it didn't look bad (not perfect, of course). I will reduce print temp as you say, to 190 then 180 and will report back. $\endgroup$ – Kevin Sep 26 '18 at 15:14
2
$\begingroup$

You can Z-hop what you like, but if it is oozing it is oozing, you will always see the effects of that as it just drops down.

Basically you have multiple issues, first the oozing, second the line markings on the top.

First

Oozing is fought by applying correct settings for e.g. print temperature, retraction, coasting, travel speed, as is explained in an answer on one of your own questions. Note that you have not explorer the coasting option as far as I understand from your question. Coasting means that you stop extruding filament prior to a move when the head is still printing. This is explained by the pressure build-up in the hotend; ideally you set the coasting length as such that all the material that is pressed out as a result of the pressure build-up is extruded just before the head moves/travels to another location. Specific calibration prints can be found to tune this for your printer.

enter image description here

Second

The markings on the top can be removed by setting the correct combing setting. Combing does not use retraction and uses straight moves, this saves a lot of time in printing, but it makes those ugly scars on bottom and top faces. You can set the option to not comb on those surfaces, and frankly, what happens on the inside stays in the inside, I would not worry about that. A good read with additional info on combing is this post.

|improve this answer|||||
$\endgroup$
  • $\begingroup$ @Oscar, you nailed the problem perfectly! Combing was the problem, for sure. The Cura default for combining was enabled; I show my updated pics above. Thank you SO much for your help and time spent making the posts. I read your link on combing and it explained it more. If you ever write a book on all this, I would buy it in an instant. Your blue image above reminds me exactly of the calibration test I printed and illustrates exactly the stringing I had between pillars. I will study coasting next. Thank you, thank you. $\endgroup$ – Kevin Sep 27 '18 at 15:56
1
$\begingroup$

Having run into this type of problem at the library makerspace, under a different slicer, I had a good idea where to start the search. It is, in your case, "z-hop cura slicer" and the best return came from Polar3D.

Keep in mind that this post is more than 2 years old, but if the feature was there before, it should be there now.

Summary (okay, dead straight copy/paste):

Cura settings > Basic > Enable Retraction - details (click the little gear/cog) > "Z hop when retracting (mm)"

You may see it default to 0.3mm, try raising this somewhere around 0.8mm - 1.0mm.

|improve this answer|||||
$\endgroup$
  • $\begingroup$ thank you for your reply. You are correct, my Cura Z Hop When Retracted setting is disabled by default. I will enable it now (it is set to 1.0mm) and report back. $\endgroup$ – Kevin Sep 26 '18 at 0:47

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.