Troubleshooting that gunk between the towers (I created this model to narrow down something I’ve seen on a few prints). Seems to be leaving some plastic behind as it moves from tower to tower, I’m guessing over-extrusion or maybe not enough retraction, anyone seen this before?
I’ve seen it. I haven’t worried overmuch about it because it only manifests in situations like those you show - multiple peaks. I suspect randomising start positions (“jitter”) might solve it.
My guess: retraction - retraction speed - temperature or a combination thereof. The hollow pyramid is also a good candidate for working these issues out. Remember though, perfection is the enemy of good enough.
Looks retraction related. As @Brian_Evans suggests, try pushing up you retraction speed and failing that, your retraction distance. These numbers can vary a lot depending on the type of drive you are using (direct or geared) the type of hot end (some are very oozy) and if you are using Bowden or not.
Lowering the temperature and/or using higher retraction distances should help. It looks like you are having some ooze that is catching on previously deposited ooze as the nozzle is going between towers.
Also, if you are over extruding, the excess plastic could end up hitching a ride on the nozzle and then getting deposited.
Traveling as fast as possible makes a big difference, though “as fast as possible” isn’t very fast on some machines. Also, if you’re using Slic3r, try the “wipe before retract” feature.
Thanks for all the great feedback!
Sounds like retraction is the place to start. I started to experiment here but had no idea what magnitude of changes were in order.
I’m using a MakerGear drive (geared) and a J-Head hot-end (modified to use QU-BD electricals). Overall I’m satisfied with the results, but I have a few anomalies like this that I’d love to tune-out.
Thanks for the mention of “wipe” @Whosa_whatsis ! I saw that checkbox but assumed it required special hardware (saw some kind of “wiper” mod on a Makerbot once).
Also thanks for the reminder about perfectio @Brian_Evans ns 
@David_Wood yeah this does seem worse now that I changed from a translucent blue to a green filament! and it’s even more extreme when I use a black spool that I’ve had for awhile.
Wipe in Slic3r 1+
So this is interesting:
https://plus.google.com/103331171897037694195/posts/LP2F6WYnFkB
I tried enabling “Wipe”, and backing retract down to 1 (the default, just to isolate variables) and what I got was these partial towers.
Additionally, the filament was backed out of the extruder when I went down to check on the printer, and it was printing air for about the last half of the model.
So I’m not sure exactly why, but it looks like turning on wipe caused the filament to get retracted a bit more extremely than I would have liked 
Either you have a negative value in “extra length on restart” that was more than you were trying to extrude between retractions and added up to enough to pull the filament all the way out, or your extruder is skipping when it tries to re-prime the nozzle after reversal (or possibly during regular extrusion, I guess).
Check that extra length setting first, then try lowering your retraction speed to see if that makes a difference. Also check the current setting for your extruder motor.
So after repeating the “ejecting filament mystery” (reverted back to previously non-ejecting settings and made sure the only thing I did was check the “wipe” box), I kinda “punted” at the suggestions of a few other Reprappers and tried slicing the model with Cura.
Other than attempting to shatter the glass on my platform (what is up with this default starting gcode: “G1 Z15.0 F{max_z_speed} ;move the platform down 15mm” ???), the results look much better, although I still would like to know why
