paulg

I have a UM2, 2 weeks now and after a great 3-4 days of full on and faultless printing, then began to suffer underextrusion as well, and like you, I did the Atomic method clean, but there was no evidence of any buildup. There have been a couple of causes though. I had one "event" like your last picture where the filament just stopped extruding after a while. Turns out that when the filament retracted (as it does regularly during the print) on one occasion it must have snapped off, rather than pulled back from the head. then it seems to have snagged when it was fed back into the head and didn't actually feed into the print head. The filament didn't grind out at all, but when I pulled it out, it was almost square across the end, and a broken, not melted surface instead of the usual taper. I doubt that this happens very often! The square end must have snagged on something as it was being pushed back into the head. So... make sure your Bowden tube is properly inserted into the head assembly, so that it can't catch on a lip around the print head. Otherwise even with normal retraction, it may get caught up as it feeds back in. Other than that, I think this is a rare event and in my case, I just reloaded the filament and all was good. The second cause was just not enough filament at times. It gave me almost missing seams on occasion as the layers were put down, and occasionally the top surfaces failed to close over. there were a couple of absolute disasters in there too where the model just came apart while printing. I increased the extrusion rate to 104% and this has settled things down. I'm not getting the "clicking" from the extruder so with this change I'm not pushing too much filament through. Some of the other forum threads discuss reducing the print speed at times too, but I haven't had to do this - yet. A couple of days of intense printing since my last failure, and all is OK for now. Hopefully you can get the same improvement too. Final thing: make sure the filament isn't snagging on something before it feeds into your UM. It can almost tie itself in knots around the filament spool at times, and your feeder isn't going to be able to pull it off the spool effectively when that happens. I suspect that's happened to me on a couple of runs - in the middle of the night of course when I wasn't able to see if this was what really happened.