I did some overhang test prints after noticing some strange behaviour with my overhangs. I used to be able to print these fine but after some time not using my printer and a repair due to a faulty stepper motor driver, I have been experiencing strange issues.
I printed with grey Velleman PLA at 190°C (70°C bed), two perimeters and 15% infill. These are my usual settings and in the past have been optimal.
I printed [color=#0000FF]this overhang test[/color] at 100 µm layer height with 3 solid lower layers and 5 solid upper layers. Between 5° and 15° overhangs look quite good. Between 20° and 35° they look really bad, with lots of drooping. 40° looks a little bit better, but still isn’t great. Then it looks perfect from 45° to 60°, after which it understandably gets worse again.
At 200 µm with 2 solid lower layers and 4 solid upper layers it looks fine up until 50-55° which is understandable.
this is quit an challenging question. For the beginning a few thoughts:
having an overhang of 30° you need to shift every outer layer of 30µm for a 100µm layer thickness. Now you don’t have an underlaying layer over the whole extrusion width. You need a certain pressure to fill the gap between the underlaying layer and the nozzle, except for the fraction in air.
How would a liquid behave in such condition? First of all it will flow to the open side. If the open side is small enough, only a small fraction will be spend for this side. If there is a large gap, most of the material will flow along this gap. These both conditions can run in a quite homogeneous way. Somewhere in between some small irregularities of the underlaying layer will generate large variations in the flow towards the open side.
That*s the model I would assume for your effects. It would be quit interesting to make a vertical cut through the problematic structures to get an impression about the form of the extruded lines. This would give some hints if the above model can be applied to your problem or not.