I'm printing with a Prusa MK3, with the following settings:

  • 3 perimeters
  • 50 % infill
  • infill overlap: 50 %

The filament is Polyalchemy emerald green (PLA). Nozzle temperature: 210 °C.

On a simple part (it's a keychain), the shell detaches if I apply a bit of force on a zone of the part that is "fragile". See picture. You might not be able to see it, but only the 2 external perimeters detach from the rest of the part. I used to print this part on another printer, and I never observed this problem.

Any idea on how to solve this problem? It seems the 3 external perimeters didn't fuse properly.

enter image description here

  • $\begingroup$ What is the printing speed? You are probably printing too fast. I've seen this happening a lot with PETG. $\endgroup$
    – 0scar
    Commented Dec 14, 2019 at 19:07
  • $\begingroup$ Not particularly high, 45 mm/s for the internal perimeters, 25 for the external ones. Infill 80. $\endgroup$
    – JPFrancoia
    Commented Dec 14, 2019 at 19:22
  • $\begingroup$ Nope, it's not the speed: I reduced by 50% the speeds above and got the same results, the part is brittle at the same place. $\endgroup$
    – JPFrancoia
    Commented Dec 14, 2019 at 21:13
  • 2
    $\begingroup$ Belts tight? Under extruding? $\endgroup$
    – 0scar
    Commented Dec 14, 2019 at 22:48
  • 2
    $\begingroup$ The overlap between infill and perimeters may be too low. Or there is underextrusion. I would try 3% higher extrusion rate. $\endgroup$
    – FarO
    Commented Dec 16, 2019 at 17:58

2 Answers 2


I ultimately concluded that the material isn't great. I printed this part with multiple other PLAs and never observed any problem. The Polyalchemy PLA looks great, but for any part that has some sort of mechanical constraint, it behaves poorly.


I am looking at your picture, and I realize that it is not only that the perimeter is detaching from the infill, but also that the perimeter is breaking.

Once the perimeter breaks, the weaker connection with the infill will surely break, too.

You could try:

  1. Using more than two perimeter layers. Go big. Try five.
  2. Add a fillet where the ring attaches so that the force is not focused on a point.
  3. Choose an infill percentage and pattern that maximizes the contact between the perimeter and the body.
  4. Change the angle on the bed at which the keychain is printed so that the infill maximally connects with the breaking point. That might make the ring be 45 degrees off the X and Y axes.
  • 1
    $\begingroup$ If the perimeter breaks, it's likely at a seam that didn't fully bond due to a retraction problem just before the wall was started, where the start of the wall was underextruded or missing entirely. Note that slicer settings usually prefer making the seam at the sharpest concave corner, which matches where the break in the picture was. $\endgroup$ Commented Aug 24, 2021 at 16:17
  • $\begingroup$ @r.. then may I assume that you would recommend turning on random seams, if the slicer supports it? Perhaps also changing the retraction settings? $\endgroup$
    – cmm
    Commented Aug 24, 2021 at 19:41
  • 1
    $\begingroup$ No, diagnosing and fixing the retraction settings or whatever it is causing the weak seam. Random seams just come out really ugly and aren't even reproducible, which is bad for getting parts you can test and improve. $\endgroup$ Commented Aug 24, 2021 at 20:18

You must log in to answer this question.

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