Random Extrusion Failure

Hi everybody, I'm relatively new to all this and have only had my printer running since the middle of January. I'm about this close to giving up on 3D printing: ||
I realize that the distance between lines changes with varying resolutions and screen sizes, but you get the point...

So I have a Prusa i3 from norcal-reprap. It seems to be of mostly acceptable quality, although the Z axis ends are warped and dont hold the linear bearings in a concentric fashion. But thats another story...Anyway, the problem is that I am RARELY able to finish a print without the extruder jamming. Randomly the extrusion will stop and the hobbed bolt will start chewing up the filament. There is no telling when this will happen. Sometimes it happens right at the beginning while printing the brim (using Cura), sometimes it fails in the middle or near the end of the print, and yesterday it failed right when it got the the bridge layer of the 50mm bridge torture test. I originally was using black ABS from matterhackers (3mm dia. by the way), and thought crappy filament was the problem. I had no evidence to back up that assumption other than the printer kept jamming, and that the ABS would not dissolve at all in acetone, even after more than a week (though it would expand and become like pudding). Their filament appeared fine visually and had an acceptable variance in diameter, but it was black and I've heard horror stories about black. I now believe the filament may have acceptable. I switched to makerfarm orange ABS ordered directly from their website, and the printer worked great for 4 prints! but then the jamming returned without me changing any settings. I was literally printing the same part over and over, hitting the print button for the same gcode file, and the thing all of a sudden strips the filament halfway into a print. I am using a clip and a bit of paper towel as a filament wipe to keep FOD from entering the nozzle (it is just brushing up against the filament and does not add any resistance to the filament being drawn into the extruder), and in addition I have a plastic bag that my jeans came in covering up the entire spool of filament to prevent dust from settling on it (its the perfect size and clear, if you order jeans online give it a shot).

While the printer was working, I had it doing perimeter at 30mm/s and infill at 60mm/s. I don't know if that is a reasonable speed or not, but it did work for a while. Temperature at first, when it was working, was either 235C or 240C. Oh right, I used ceramic non conductive CPU heatsink compound to cement the thermistor into the nozzle, and did PID autotune (sprinter). The heater resistor is wrapped in a single layer of aluminum foil for a snug fit. Extruder temperature appeared to be stable during prints, usually within 1C but sometimes straying 2C high or low. I tried lowering the temperature to 230C, and it failed. I tried printing a part at 248C, and it failed. Everything in between fails. My man parts convinced me to crank that bad boy up to 260C (!!) but within a minute I came to my senses and lowered it to 248C, and later in the print it failed. I have verified in my firmware that the upper temperature limit is 275C, so there is no way the extruder should be doing the emergency shutdown thing. I have selected the correct thermistor in the firmware. My esteps in the firmware are calibrated to within half a mm over a distance of 50mm, which should be close enough to fine tune with the extrusion multiplier. Using M105 I have verified at the first sign of failure that the temperature is as I had specified in the gcode, and it is. I don't have a way to verify the temperature with a thermocouple, but I did put a wee bit of orange ABS on the side of the nozzle, and over time it has become black, so I think the extruder is hot enough that it should be extruding. The plastic does not come out looking watery, so I don't think it is too hot either. Air extruding results in a nice shiny 0.5mm diameter noodle (this is a 0.4mm nozzle), and it seems to extrude nearly straight down. The noodle is circular, not oval shaped (confirmed with calipers).

The nozzle doesn't seem to entirely plug up as I would expect. If I pause the print, pull the filament out and clip off the chewed up portion, and put it back together it will continue printing happily for a while (usually tens of minutes) before jamming again. I've actually found that no matter if I clean the nozzle out or not, it inevitably fails at some point in the near future, and it doesn't particularly seem to matter if I simply clip the chewed up part and immediately continue, or spend a bunch of time trying to clean out the plastic thats already in the extruder. I can keep trimming off filament and reloading it when it jams every so often, and it does actually make it through a print (albeit with defects from where it was paused for 30 seconds while I fixed the extruder). I have tried the suggested method of cooling the nozzle to 140C and reversing the filament to pull all the plastic out, but so far have found this to not work quite as advertised. My best result was at 135C, and a reverse speed of 30mm/min (slower than I can go by hand), but the filament still ended up snapping with a bit left in the extruder. It did make a very long stringy thing and I'm sure pulled out most of the plastic with it, but there was still some left. A hotter temperature will melt the filament in half, and a lower temperature snaps sooner. I repeated this process 5 times in a row, and afterwards it still failed to print a complete part without jamming.

I have tried lowering the print speed to 15mm/s perimeter, 30mm/s infill, and it still fails. I suppose I could try going slower, but I hear about people going 80 or 90 mm/s and I see no reason why I should have to go below 30. Is 30mm/s unrealistic?

The hobbed bolt appears to be of very high quality. The teeth are evenly spaced and the groove does not wander back and forth as the bolt rotates. I clean the teeth out each time it strips the filament. I have gone so far as using index cards as washers to center the hobbed bolt over the extruder opening. I have tried adjusting the extruder spring tension bolts from extremely loose, to absurdly tight completely compressing and crumpling the spring coils over top of eachother, and it still fails. I have listened to the stepper motor and it is not skipping steps. I have checked that the extruder gears are not slipping on the motor shaft or hobbed bolt (but it is chewing filament so you already knew thats not the problem). I have checked that the filament can be pushed through to the bottom of the extruder, and while it cannot go ALL the way, it is fairly close I think. I believe the last bit of distance is because there was still some small quantity of filament in there. It certainly goes down past the top of the brass and into the melt zone. I find it fairly difficult to extrude by hand, but the motor seems to have no trouble. That is, until it randomly strips the filament again. I currently have it set up to slightly underextrude so that I can be certain it is not overfilling plastic and plugging up the nozzle by dragging through previous layers. Solid layers have a very slight gap between the infill.

In case I missed anything, here is a summary of the printer: it is fitted with the hinged accessible extruder, guidler, and a genuine J-head from hotends.com (I know it is genuine because I broke the PEEK on the original one trying to disassemble it while it was cold, and bought an entire new J-head from Brian). Electronics are sanguinololu with pololu drivers and sprinter firmware. Firmware max temperature set at 275C. Thermistor cemented in using electrically nonconductive ceramic heatsink compound. Heater resistor wrapped in aluminum foil for a very tight fit. Filament was originally black 3mm ABS from matterhackers, and is now orange 3mm ABS from makerfarm, both of which I am experiencing these jamming issues with. Filament roll is covered with a bag to keep dust off, and it passes through a paper towel clipped around the filament immediately above the extruder. It is very very easy to slide the paper towel clip thing back and forth on the filament, so I am 100% certain any resistance added by it is negligable....that's about all I can think of.

So seriously, WTF is wrong with this thing. I've been fighting with it several hours each day, and practically all waking hours on the weekends, for nearly two months now. I'm about to throw most of it in the trash and use the steppers and electronics for other stuff. If anyone has ANY ideas, please don't hesitate, because at this point I'm completely out of ideas to try other than lowering print speed to a snails pace.

Its purely the heating block,

Please go step by step, you can achieve.

1) Even before extrusion, please ensure your Extrusion drive is proper. whether is it pulling filament properly.

2) If you are using Slicr, please use the basic settings right.

3) Jus ensure your extruder motor shaft and the gear is properly connected, and rotates along each, sometimes the shaft alone can rotate or might be loose, so it might pull the filament and leave it.

4) Use a controlled environment, don use the extruder cooling FAN for testing.

5) Do a Extrusion calibration with proper extrusion multiplier values.

6) Dont enable any advanced settings in Slicr or dont tweak the retraction settings on Slicr.

With this all you should achieve, by the way, can you post some pictures to know about your extruder and hot end.

Thanks for your response.

1) As stated in my original post, the extruder works fine for a while. It feeds filament properly at the beginning of a print. It does not skip steps.

2) As parenthetically stated in my original post, I am using Cura. I have all the options set appropriately as far as I know. I intentionally slightly underfill the print just to be sure the nozzle does not drag through previous layers.

3) As stated in my original post the gears are not slipping on the shaft, and the stepper is not overheating.

4) The printer is sitting in my office, which is heated/air conditioned depending on the season. The temperature remains relatively stable. It is not directly underneath a vent or subjected to other airflow. I do not have a fan for the extruder.

5) I'm not quite sure what this refers to. I do have esteps calibrated properly as mentioned in the original post. I have the extrusion multiplayer set to 100% in Cura but have typed in a larger filament diameter than it really is in order to have slight underextrusion.

6) Again I'm using Cura, and the only "advanced" feature I have enabled is brim, which I find necessary in order for the part to stick to the bed. I have retract set at 30mm/s and a distance of 0.75mm, but Cura is weird and doesn't do retracts unless you print multiple parts. Normally it does not retract and moves the nozzle to keep it inside the part. Any stringing gets hidden in the infill. I don't think there is anything else set up that would mess up the extrusion. The temperature remains constant. There is no option for line thickness, only nozzle diameter, and it is set up to under extrude, so I don't think its trying to force too much plastic out the nozzle. Oh I set the infill/perimeter overlap to 0 just to make sure the nozzle doesn't drag through any plastic. If anything that should make it less likely for the nozzle to plug up.

Again thanks for your input. I'll download Slicer and set it up as you have suggested. I have a feeling that switching to slicer will not cure the extruder jamming problem though.

Also I'm at work otherwise I would have taken pictures of the extruder and attached them to this post. I'll attach pictures with my next post later tonight.

SniporBob:

I am afraid that I might not be able to help but did want to confirm that you have never extruded PLA from the troubled nozzle? I mixed PLA and ABS and think that resulted in a clog to beat all clogs. Couple of other questions, have you tried white ABS? My understanding is that black is the most troublesome of ABS (not sure about Orange), maybe white would be more friendly. Lastly, I struggled with ABS and will come back to it but have had much better success with PLA (now that I have dedicated a hot-end to PLA). Maybe you could try PLA and get some success and then move to ABS (at the cost of another hot end)?

Wisar

I had a similar issue when I first put together my printer and the cause was that the grub screw at the top of the hotend that was holding the PTFE liner in place wasn't on properly, so the pressure inside the hot end was forcing the molten plastic in the small gap between the PTFE liner and the heater block, forcing it upwards and forming a slug of semi-molten plastic that caused a blockage. Here's a photo of a disassembled J-head:



Perhaps you could disassembly your hot end, make sure there are no slugs of plastic inside and that the PTFE liner is sat right down against the heater block with the grub screw firmly keeping it in place. If you do disassemble it, don't forget to put some sort of thread sealant on the heater block when you put it back in otherwise you might get molten plastic leaking out.

Wisar: an excellent question regarding the PLA. I have not run PLA through this hot end ever. This particular hotend has seen maybe half a meter of black ABS, and quite a few meters of orange ABS. I don't have white but perhaps I will order some in the name of science. I take it white ABS is regarded as the most trouble free?

I initially struggled with the decision to print ABS or PLA. I ended up deciding on ABS because it actually seemed more people were having issues with PLA than ABS. Maybe that perception was incorrect...?

Amrbekhit: I got the hot end from hotends.com so I had assumed (yes I know...) that the PTFE setscrew would be tightened properly. That is worth checking though. I will disassemble the extruder tonight and report my findings.

Side note to anybody who hasn't disassembled a J head before: You must do so with the J head hot. I tried to take my first J head apart cold (because I didn't know better) and the PEEK insulator split in half vertically.

SniporBob:

I am no expert but did read that black was particularly bad with impurities. I had also read about people having issues with PLA but it has been much easier for me to work with. Less warping and bed adherence issues and the heat being lower means that I spend less time waiting for stuff to reach temps. Ironically I am having an extrusion issue as we speak but it is the first with PLA after printing nearly three spools of filament, and, I think it is actually a servo issue where the z axis is not lifting thereby causing back pressure and the resultant clog.

[Update] Was just a simple nozzle clog. Used .4mm drill bit to clear. All is well here. Good luck there...

Wisar

Edited 1 time(s). Last edit at 03/12/2014 10:41AM by Wisar.

Quote
sniporbob
I initially struggled with the decision to print ABS or PLA. I ended up deciding on ABS because it actually seemed more people were having issues with PLA than ABS. Maybe that perception was incorrect...?

Yes, your above perception is incorrect... PLA right off the top is easier to print with, doesn't stress many extruder hot ends and print beds with higher print temperatures, easier to control geometry and dimensional size (ABS is a shrinking, warping, dimensional nightmare, constant battle to compensate for it's ill behavior).

The early stoneage days of RepRap... Spindly M8 Threaded Rod Chassis, Wades Geared Extruder, Hobbed Bolts, Printing with ABS and using 3mm Filament for print sizes <200mm x 200mm.

All I can say is ditch the ABS... print with PLA... and kiss warping, deformation, and ABS headaches good bye, and never look back.

BTW... Black filament whether it is ABS or PLA is seemingly proven to be the most difficult of all to print for most users.

Hope my experience helps you... enjoy Happy Printing !

<sub>Shawn
__________________________________________________________________________________________________________________

25+ Years as a Manufacturing Engineer and New Product Developer
Three Years Experience... Printing 6-8 hours a day, 5 days a week, complex and close tolerance functional engineering prototypes.
Printing PLA @0.1mm Layer Height with +- 0.05mm tolerance for all dimensional shapes (except slightly undersized hole I.D.s).
No printing of... toys, trinkets, low quality prints, or other useless printed objects.</sub>

Edited 1 time(s). Last edit at 03/12/2014 01:53PM by ShawnT98027.

Quote
ShawnT98027
__________________________________________________________________________________________________________________

25+ Years as a Manufacturing Engineer and New Product Developer
Three Years Experience... Printing 6-8 hours a day, 5 days a week, complex and close tolerance functional engineering prototypes.
Printing PLA @0.1mm Layer Height with +- 0.05mm tolerance for all dimensional shapes (except slightly undersized hole I.D.s).
No printing of... toys, trinkets, low quality prints, or other useless printed objects.[/center][/sub]

Guess that makes the rest of us, idiots then.

---

_______________________________________
Waitaki 3D Printer

Sorry for the lack of updates everybody. The good news is I managed to complete several prints without a jam! The bad news is I had to slow the printer down to 20mm/s perimeter, 30mm/s infill. Tenp was set to 230C. When I slowed it down, I found it to be over extruding. If it was slightly under extruding when going faster, and is now over extruding when going slower, my thinking is that the extruder is unable to keep up at the higher (although still slow relative to what everyone else seems to print at) print speed. I may just resign myself to these slower speeds.

Oh right, so before I slowed it down and got it to work right, I took the nozzle apart. It seemed the setscrew was tightened down well on the PFTE. I don't know how tight it is supposed to be, but it wasn't loose and it wasn't super tight either. There was not any plastic buildup in the PTFE. Its a very snug fit on the filament so I don't see how that would even be possible. The plastic in the brass nozzle didn't look burned or anything. I did notice a little bit of black stuff in the threads of the PEEK, so I picked that out with a pliers. There was no thread sealant used on my nozzle. There werent any big signs of plastic leaking out between the PTFE amd nozzle. Seeing nothing obviously wrong with it, I reassembled it the way it was except I did add 3 or 4 turns of PTFE tape around the nozzle threads. I slowed down the speeds to what I mentioned above, hit print, and found that after a short time some black stuff started coming out of the nozzle mixed in with the orange. I was very concerned that I had somehow messed up the PEEK, but let it keep printing. Eventually it returned to pure orange after a layer or two. It has been working fine, albeit slow, ever since.

Now some questions.
1) What would normally be considered a reasonable print speed for ABS (if my machine were working properly).
2) Is the PTFE liner supposed to fit snugly around the filament, or should there be a little extra room to allow for filament expansion? Or does the PTFE expand more than ABS?


I'll have to give PLA a shot sometime. I did read a lot about people having problems with it buckling in or above the hot end, which scared me away. I don't seem to read nearly as much about extrusion issues with ABS, but do come across lots of warping stories. I don't get much warping anymore but I do have issues with overhangs curling upwards. That's for another topic though.

Snipor:

It is a shame no one answered your last couple of questions. I took a break from printing PLA, and tried a new ABS hot end. I am getting jams just as you describe. Am back to PLA until I get a chance for some more testing.

Wisar