For the outline, the end mill wasn't low enough, so I had to pause (let's say, at z=1), and then lower it to the suppressed surface (to z=-1), and then ran it again without setting it as new z0. 
It means, the machine considered the current z=-1 as z=1 (which is lower by 2 now), and when it touched the surface thinking it's z=0, it was actually z=-2!!!! 
 
So, the end mill was stuck and then the upper connecting part had to spin in reverse direction, unscrewing itself. I might have damaged the machine.
 
 

Go back