Nicolinux

Hey you're right. The knurled bolt mantle is not centered on the filament. The filament sits on its edge. It could be that it doesn't always bite correctly. Ahh, I don't want to disassemble it again. The setscrew of the knurled bolt mantle is damaged. I might not be able to tighten it again properly. I think I'll order one on Ebay first. But thanks for the hint. This could be it.

I'd say there are different issues then. I do get underextrusion even at relatively low speeds. But my issue might not be related to the extruder after all. Can't wait to test the new teflon coupler. By the way, I ordered a digital scale that can take up to 10kg. My next test should be way more accurate.

Hey Siggi, Thanks for the idea. Already checked it. I put a mark on the stepper shaft and the knurled bolt mantle. The didn't move a notch since then, so I assume it doesen't move/slip.

About the support structure - check out Meshmixer. Illuminarti has a great blog post about it (http://www.extrudable.me/2013/12/28/meshmixer-2-0-best-newcomer-in-a-supporting-role/).

@gr5: Thanks for the effort, I'll repeat the test. Though I have a digital scale that can handle only 5kg. I'll try it - don't trust that analog scale that much. @braddock: I think the reason is because there aren't many users with this issue. Or at least they may not be aware of it because they print slow. Maybe one of the forum admins could make a sticky post about underextrusion and redirect people here and to my underextrusion monster thread.

@AaronAlai: Hehe, call it a doodle and you might get away with it as "improvised art" @gr5: Thanks for the update, will test again.

I can't measure the temperature inside the teflon/nozzle connection. There is no way to reach in there even with the tiny temperature sensor I have here. I measured the temperature outside, around the height where the coupler and the nozzle should meet. It is about 110°. So PLA will definitelly expand if there is space left. Haha, mee too I moved my UM2 to the living room because my Ultimaker happy corner is too small for tinkering. Good thing there are no ladies around right now, the WAF (woman acceptance factor) is way too low around here. By the way, printing the retraction variant didn't prove the theory. It even fared better than without retraction. Also the object on the left was at the front and features way more underextrusion than the on in the back. But I think that's just coincidence.

Wow, thanks for the trouble you went through to test this. Very impressed. And you are right, I think this could happen. I mocked-up this scenario from the image you linked: I imagine the grey filament gets caught in the space between the teflon exit (blue ring) and the nozzle entry (purple ring). It forms a bulge and makes it harder for the extruder to push through. And this is especially troublesome if retraction is involved. To test this asumption I will print the retraction version and see if it fails earlier than the no-retraction version.

Cool print (nice color choice). And this grungy hardware backdrop fits very well.

Yeah - we are going to crack 20 pages soon! :eek: I have loosened the screws on the extruder. Didn't help at first, but after I loosened them even more it helped a bit. Got to a point where I feared the stepper would fall off, but the extrusion test still failed at 6-7mm^3/s And another peculiar thing I noticed. Don't know how it could be related though. The teeth marks on the filament inside the bowden tube do not form a straight line but somehow rotate around the filament. I haven't noticed it before, maybe it is normal (and due to the filament curvature).

Thanks!

Ok, thanks. Good idea, I'll try it. It would fit because I tend to over-tight things. Happened to me with the UM1 too. By the way, I tried a makeshift filament guide so it enters the extruder straight from the bottom. Didn't work, so that's not it either. I am just documenting all failed attempts here in case there are others following along with the same problem.

Thanks for the links. But I don't know... Already spent so much time (and a bit of money) on this issue. I will pursue one more idea until the coupler arrives (looking to build a filament re-routing guide right now).

Hey guys, I just stumbled over this post from Daid where he statest that the UM2 extruder motor is stronger than the UM1 (http://umforum.ultimaker.com/index.php?/topic/4197-improving-the-um2-feeder-mechanism-for-flexible-filament/?p=35682). Now, why would the UM2 pull less than the UM1?

Hi, Thanks for the hint. I'll check the extruder entry hole. Sadly I don't have a spare bowden tube, but I'll try to re-route the filament by other means. On another note, this weak extruder of mine is still bugging me. Provided that I measured accurately, 30% difference is a lot. I'd like to understand (and fix) it. If it is a normal manufacturing variation I wonder which factors could facilitate such a big difference. Given that all components are identical, only the stepper driver config could be it - right? One thing I noticed (I removed the cover for the extruder stepper) - during and after a print the extruder stepper is fairly cool compared to the other steppers. I'd love to have some documentation on this, but UM is taking its sweet time (or I didn't find it).

Another update. I disassembled the head and enlarged the exit of the teflon coupler. I made sure there is no lip anymore. Ran the extrusion test again after assembly ...and ...nothing changed :angry: Still fails right after 4mm^3/s. Sander offered to send me a new teflon coupler. I'll try it as soon as it arrives. I hope it works, because that's one of the last things I'll try with this printer. On the bright side - I am now fairly familiar with the head/nozzle and can take this thing apart and put it together in a few minutes. You know, army drill style when cleaning weapons

Finally some normal persons

I have printed Robert's awesome low friction spool holder (http://www.thingiverse.com/thing:235925). It really makes the reel rotation very smooth. Printed another test object and it fails early. So this time I am pretty sure it is not the weak extruder. I will focus on the head/nozzle again.

Hey jjshortcut, we might have the same problem. Might I point you to my monster thread where many people contributet quite some ideas? (http://umforum.ultimaker.com/index.php?/topic/3976-almost-always-missing-layers-underextruding/). Another issue could be the teflon coupler that sits on the nozzle. Have you checked that? You say that you can barely push filament by hand which means the problem must be further up the bowden tube (head probably).

Hey guys, another idea. When you start Cura for the first time, I think it asks you if you want to send data to Ultimaker. If it does, and the people who bought one of the stolen unites performs a firmware update - are you able to catch the serial number?

Hey guys, just a wild idea here. Wouldn't it be possible to measure the voltage of the stepper and extrapolate the actual torque? If it is possible, that no additional hardware is required. I found this discussion on the Arduino forums. Maybe it is useful (http://forum.arduino.cc/index.php?PHPSESSID=hthtsj0ddum9ceufng9j53ddr3&topic=136261.msg1024483#msg1024483).

Aha! Thanks, I am not crazy after all

I am printing all test objects at 230°. The thing is, I am not printing test objects day in day out just to prove some fabrication tolerances for the UM2. I do print stuff I want and most of the time I stay well under 4mm^3/s and still get underextrusion. Take this vase for example. Printed at 0.2 layer height, 40mm/s speed (3.2mm^3/s) and 230°: Some prints are taking forever and since I live in an apartment I can't afford to print overnight because I am a nice guy and don't want to disturb my neighbors. So I am trying to balance speed and quality. Besides, I see that other UM2 owners can achieve what I am looking for - printing with the same settings (illuminarti for example). I am not trying to optimize the last 5%, I just want to get rid of underextrusion. And I don't think that going almost max temperature is a solution either because this has problems on its own. EDIT: Another note. This underextrusion problem creeped in over time. I printed objects earlier without underextrusion. Maybe I got lucky and used some settings that didn't favor this problem, but if I remember well, the underextrusion issue got worse over time.

No wait, that's just heavily underextruded PLA, pressed down to a fluffy heap

Fed in the standard way. Spool on the spool holder and filament fed in following its natural curvature.