# Infill failure at the far corner of two different prints

I am a complete newbie at 3D printing (although learning fast). I switched from the default Qidi slicer to the Cura slicer on a Qidi X-Pro machine (which is wonderful). Using: plate 50 °C, 0.4 mm nozzle 200 °C, 1.75 mm PLA, print speed 20 mm/s outer wall, 40 mm/s inner wall, travel speed 100 mm/s (the cura default, I think).

On two different prints, the same far corner (front left) had initial infill problems. Does anyone know why, or how I can solve it? My working theory is that the nozzle shoots across the whole print at high speed to start again at that corner and sometimes the infill filament doesn't bite enough to stick.

Eventually one of the infill layers gets a bite and the rest of the print is fine. But it's disturbing that it occurs on different prints. (I carefully level the plate before every print, so I don't think that's the issue.)

UPDATE

In another question here, turning the Cura combing setting off solved an issue. It probably solved the issue described here, too, or at least contributed to solving the problem described here.

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 corner to bite and bond to the corner wall. Turning the combing setting off seems to have solved the missing infills in the corner. It also greatly improved the quality of the infill walls, too, which relates to the problem described in this question.

UPDATE: Here's a paragraph from a forum posting here that explains how oozing caused by combing can result in "underextrusion effects." That's what I think is happening in this post - the nozzle gets to the corner, but it has no filament to bite and bond with the wall. Credit to @0scar for giving me the forum link. Quote from the link:

Even on interior layers, combing as implemented in Cura is a bad idea, as it doesn't perform retraction, and that can result in the head oozing as it moves, especially as each time the head crosses a line of infill it can tend to pull our some plastic. You can also get some long moves (e.g., all the way around the curve of a 'C' shaped object when moving from one side of the opening to the other). This can result in the head being empty when it starts printing the perimeter again, with resultant under-extrusion effects.

Now that you are using Ultimaker Cura, the default (probably hidden) parameter Z Seam Alignment is set at Sharpest Corner, so it will always start in a sharp corner. The default retraction settings are pretty high in Ultimaker Cura (I think about 6.5 mm as default Retraction Distance). What happens here is that the filament is not available for printing in time (this is dependent on the extruder type), so the distance should be less than the current value.

It is highly recommended to print a retraction test to find what is best for your setup. For me the 6.5 mm works very well on my Ultimaker 3 Extended and my large custom CoreXY printer as they both use a Bowden extruder setup using 2.85 mm filament, this can be different for your printer. Choosing a different Z Seam Alignment option, e.g. random prevents the use of the same location at every height, but could result in a less aesthetic print.

Furthermore, you could try to visualize the G-code in a viewer, this can be done in the Ultimaker Cura slicer itself, but online viewers are available. After slicing your product, change the combo drop down box from Solid view to Layer view to show the sliced object. Nowadays there are 2 sliders to interface with the model; the vertical sets the layer, the horizontal the progress of the layer. It is advised to play with this to understand how your print is actually printed.

About your settings, print speeds are not high (could be increased), travel speed is fine (is only for non-printing moves). The brim is looking good and the level of your bed is also good.

• Hi Oscar, I checked and you are correct! The default retraction was 6.5mm, so I reduced it to 4.0mm. I will report back after my next print. Should I change z-seam alignment to Shortest or Random for better results? Thank you – Kevin Sep 18 '18 at 20:04
• PS. If the problem is caused by the filament not being ready enough to bite in the corner, why isn't Cura smart enough to push the filament out the right amount in the right amount of time? It knows that it retracts 6.5mm, so shouldn't it extrude 6.5mm (or whatever is best) to get a bite? I really appreciate your help. It sounds like you should have a book out that I could buy too... :-) – Kevin Sep 18 '18 at 20:08
• @Kevin, the problem is that right after the +6.5, the filament is melting and being forced through the nozzle. This takes a small amount of time, and is exacerbated by any loss to oozing at the start of the layer change. These are really small effects, but enough to affect some prints. – Sean Houlihane Sep 20 '18 at 11:46
• @Oscar, it looks like you updated your answer with a retraction test. Thank you - I didn't even know there was such a thing. Perhaps you could point me to a checklist of all such good tests to set up my printer? Does such a list exist? Sean, thank you for your reply too. I thought for the corner problem I would reduce travel speed so that the head allowed more time to bond before taking off. Also I reduced retraction a bit to give more output in the corner (didn't help much, IMHO), and chose "Random" for starting each layer so the starts didn't add up in the same corner. – Kevin Sep 21 '18 at 15:47
• I still have some hairs and strings (esp between short 5mm support sequences on the build plate), but now at least I know about retraction tests! I will do my homework and report back. Thank you both again. – Kevin Sep 21 '18 at 15:48