How to avoid this kind of defect ?

I'm very satisfied by my M200 but sometimes there is this annoying defect and I really do not know how to avoid it.

As you can see the upper part is perfect while the bottom part not.

Every suggestion is welcome .

Reduce the fillet radius.

Keep support turned on so that even a small angle will have support, this will form a "wall" that will keep the unattached thread from stretching across the corner, it will waste a small amount of plastic but giver you more success.

All that said, this will continue to be a problem and there are some great explanations on the web why this is also the cause of "layer de-lamination" in corners on tall, large parts.

I'm very satisfied by my M200 but sometimes there is this annoying defect and I really do not know how to avoid it.

As you can see the upper part is perfect while the bottom part not.

Every suggestion is welcome .

Which material, and what layer thickness, are you using? It's not sticking to itself very well.

I don't see how Drew's suggestion of support can help, since this is the bottom of the part.

Thanks for replies !

Julia, the material is  Z-Abs , layer 0,14

YosemiteSam, you mean the fillet radius of the part ?

Drew, support was turned on, but it was not generated .

The problem occured only on one bottom side of the part , the other side is ok.

Did you use the high speed setting?  

I know I only had that problem with the high speed setting. Also what was the calibration on the area of the bed? Kinda thinking a different bed placement of the object may get rid of that too. 

Hi HeavenlyCreation,

speed was normal.

I think that bed  calibration is ok.

Drew, support was turned on, but it was not generated .

The problem occured only on one bottom side of the part , the other side is ok.

If you up the degree for support generation, to say "Angle 80 degrees" then the support can act as a guide to keep the stringing more in line on the bottom of all surfaces.

Thanks for the suggestion Drew, I will try next time I will have to do this  print .

The print was almost the size of the bed plate in diagonal .

As the extruder is doing those profiles there is only a little bit of the strand hanging on (since its an overhang) and the strand is also under tension from the nozzle; the combination can lead to the corner strip being pulled off. The way I have solved this in the past is by doing either thing Drew said (reduce radius, add support), as well as turning off 'fast'. 'Fast' shouldn't make a difference in theory, but I think they have they extrusion-to-speed ratio a little higher on regular which leads to less tension on the strand.  

Maybe submit this one to bugtracker?

I've had this happen as well in the past, not as bad as this though.