RE: Filamentum PLA Cristal Clear: Mid-Print no more layers
Hi cevans9921,
thanks for your input.
What would be (according to your knowledge) the typical flow rate value or difference value between a filament 1.75 +/- 0.05 compared to 1.75 +/-0.02mm?
Thanks in advance!
Gerardo
RE: Filamentum PLA Cristal Clear: Mid-Print no more layers
the values will be the same since both are 1.75 nominal dimension - the issue will be slight over/under extrusion ( varying extrusion width) as the filament diameter varies
RE: Filamentum PLA Cristal Clear: Mid-Print no more layers
Dear @david-a66!
the issue will be slight over/under extrusion ( varying extrusion width) as the filament diameter varies
Definitivly not. Variation in filament diameter are random and would not lead to exactly reproductable results.
Cheers,
Volker
RE: Filamentum PLA Cristal Clear: Mid-Print no more layers
@volker-jaenisch
I couldnt say on these machines. I'm used to 3D printers that I have to manually calibrate my self. Usually I have to measure the distance traveled or in your case measure the amount of filament that is being used. So if I move 50mm in the software and I get 40mm of plastic extrusion Id have to do the formula to adjust it to the correct steps per minute. These prusa machines calibrate every thing for you which is typically a lot better unless you run in to a problem like you are now cause you cant just go and edit the steps per minute profile sense it would change the rate across every thing on this machine.
RE: Filamentum PLA Cristal Clear: Mid-Print no more layers
@cevans9921
I am still not convinced that the problem is caused by filament diameter variation. Also I do not like to start a discussion if Prusa is the best printer. For me it is the only printer I have and I like to master it. Also I do not agree with :
unless you run in to a problem like you are now cause you cant just go and edit the steps per minute profile sense it would change the rate across every thing on this machine.
The Prusa is just a robot, and it followes the gcode it is given. If you meant the Prusa-Slicer then I agree with you that there is a lot of cool optimization at work. But the prusa-slicer has so many parameters, that I am pretty sure I will find the right one to tweak when I have found the cause of the problem.
As long as there are no frequent retractions e.g. due to 100% infill resulting in arcs the print runs fine. But I am a software developer and I like to do it the hard way. I like to understand any bit of the machine I am programing/using under the most inconvenient circumstances. Without this I will not trust the machine/software and cannot sleep well.
Today I made three prints:
1) An wedge test-cone.stl shaped dome where the slicer cannot translate the infill into arcs. Bad case.
This is a simpler and way not so time consuming print
2) than the original print: A hexapod ladybug mobile home robot, designed and created by my dougther, for her arts exam.
She wants to print the upper shell in transparent blue, mimicking translucent solar panels to power the robot.
3) The 8mm upper part of the wedge.
Printing the upper shell of the robot results in extruder jam at 3 cm height. The transparent lower cockpit glass that I mastered already is way higher than the fail height for the Filamantum PLA Transparent Ice-Blue. I do not like to open up a new thread since I changed the filament from one color to an other since the effect is still the same.
Printing the wedge the problem can be seen at its top, but at a greater height. If I print only top of the wedge the problem does not occur.
So the problem has still some degrees of freedom:
1) It may be a height related (aka mechanical/friction) problem.
2) It may be a temperature related problem determined by the working time.
To rule out the enclosement and its interaction with the tubes, it printed some new hinges for the MMU so that it sits a bit lower to allow for more Z-headroom. https://www.thingiverse.com/thing:4057710
Also I build a simple webcam utilizing a Raspery-Pi-Zero-W to watch the progress. https://www.thingiverse.com/thing:4057725
The next shape I will print is the top of the wedge that is elevated by a simple block. Printing blocks was no problem with this filament for at least 60 mm.
So I can check if the problem is height related.
Also I will print lots of the small 8mm wedges in an array to check if the problem is time/temperature related.
Please stay tuned
Volker
RE: Filamentum PLA Cristal Clear: Mid-Print no more layers
It is a time/temperature problem!
Printing several wedges leads to a problem at the third wedge. So it is no height/mechanical problem.
There are three factors that may induce heat:
1) the heatbed/enclosure
2) the extruder cooling fan (by not cooling enaugth)
3) the extruder motor
I recently studied all of these effects, with no outcome. So I have to go into them once more. But now I have at least a simple base case for testing.
Volker
RE: Filamentum PLA Cristal Clear: Mid-Print no more layers
Testing for heat problems:
Activated active cooling of the stepper motor. Placed a powerful 12V fan with case on the stepper.
The outcome is worse than without: The second wedge is spoiled.
RE: Filamentum PLA Cristal Clear: Mid-Print no more layers
have you physically felt the extruder when it happens to see if its any hotter then normal? stick your finger in on the side by the pinda probe and see if the heating block of the extruder is any hotter then when you started the print.
RE: Filamentum PLA Cristal Clear: Mid-Print no more layers
Hello,
this is a well known issue, and I explain somewhere here in a post with detailed documents and drawings.
In short, PLA are not equal, and the temperature at they change their molecular state from solid cristal to amorph is different. The culprit of the issue is the MMU2 PRUSA version heatbreak in the extruder, that is differenet that the E3D version.
While the E3D version is just cylindrical , and the filament moves freely all at the same way, the PRUSA one, is hourglass shaped 🙁 . They made this change to simplyfy the MMU filament changes.
However this “improvement” added another issue: when the temperature raise along the metal nozzle to the heatbreak, some PLA start change their molecular state early than others, starting to became flexible and moldable (very little, but enough) during several heavy duty high flow movements (like infill), and they assume the shape of the high larger part of the “hourglass” shape.
This larger diameter part is not able to pass through the small diameter just below, but since the filament is still amorph but not fluid, with a huge effort of the motor, it is pushed in the small diameter section, and then completelly melted in the nozzle. Here is why you get the motor hot, due to the extra effort done.
As soon as, for any reason, the metal temperature slide change and become lower, this large section become cristallized and solid again, then it stops to pass through the hourglass tube 😱 the motor become very hot, you hear the click of the filament shifting, and your print is GONE 😰
After a lot of study I solved the issue by replacing the PRUSA Heatbreak with the E3D original heatbreak that has all the same diameter.
Look around my post in the forum, and you could find all the detailed drawing scheme and picture related to the issue.
Regards
RE: Filamentum PLA Cristal Clear: Mid-Print no more layers
Well, after lots of research around the special Prusa heatbreak and watching several youtube videos about the potential jamming problems that may be a result of the stepped opening inside, I bit the bullet and replaced mine with a genuine E3D heatbreak. These do not have a step for the filament to jam in when printing with lots of retractions. The image shows a print I could not get to complete successfully prior to the swap. This print would always fail midway through printing the fine details that make up the side vent when I tried to use this transparent purple filament.
While I'm not calling my problem solved, this is very promising for my setup.
RE: Filamentum PLA Cristal Clear: Mid-Print no more layers
Dear all,
thank very much for your comments and help to clarify this issue. I guess and hope that antimix and stevensmith are talking about the hot end PTFE tube and not about the hot end heat break or the extruder E3D itself.
I just spend some time reviewing the different types of hotend PTFE tubes and was surprised by the amount of them. I wonder why aren't there any part numbers available. This would help to know exactly which part (version) we are talking about.
The hotend PTFE tube had originally a length of 50mm. After the modification of the extruder (MK3 --> MK3S), the length was reduced to 44.3mm. But the mayor modifications are related with the inner diameter and outer geometry (chamfers). The outer diameter had been kept to 4mm. The actual version that should be installed now by default should be this one (A):
In my case before installing the MMU2S I had this PTFE tube version installed (B).
Originally with the printer (MK3 R3) I had this one installed (C):
So, as you can see the mayor modifications are the diameter 1,85 vs 2.00 mm, the front chamfer and the 2.3mm counter chamfer in the front area.
I guess that stevensmith means that he replaced (A) with (B). As you can notice the counter chamfer in the front is designed to fit the filament diameter in the retract or unload direction. If in this area is too much friction or resistance (because the filament is been forced to change his diameter from 2.3 to 1.85mm) the extruder stepper motor will become hotter than usual. My guess is that with some filaments, the compression resistance builds up to such a level, that it ends up in a filament jam. Which could explain why the phenomena is related with the filament retraction amount of moves.
HOWEVER, I believe that this is a part of the puzzle, but not the final explanation. I had filament jams also with (C) where no counter chamfer is in the front area.
And yes, PLA is not equal PLA. The molecular structure, the length and distribution of the chains, the type and amount of additives make polymers more or less very unique and difficult to compare with each other even if they have the same repetitive units and belong therefore to the same group or class.
RE: Filamentum PLA Cristal Clear: Mid-Print no more layers
@gerardo-w
No, I was not referring to the PTFE tube. I swapped the actual heat break.
has a diagram of the actual Prusa specific heat break that shows the step that goes from 2.2mm to 2mm. It also shows the type of jam I was experiencing with prints that needed a lot of retractions. I would get a hardened blob of filament that was 2.2mm in diameter that would be 'trapped' between the 2mm hole for the nozzle and the 2mm hole of the PTFE tube. The E3D heat break is 2mm all the way through.
BTW, that is the same fix that the 3D Printing Nerd ended up doing on his MK3 in one of his videos.
RE: Filamentum PLA Cristal Clear: Mid-Print no more layers
thank you. This makes sense and explains finally the jam phenomena to me.
From the beginning (MK3) I was irritated by the bulky non cylindrical shape of the filament tips when they were unloaded. When the R3 modified parts came out, my jamming issue persisted. Also when upgrading to MK3S the problem persisted with the same particular PLA filament. Now I finally understand why this happens.
Thanks for shearing your knowledge and experience here!
RE: Filamentum PLA Cristal Clear: Mid-Print no more layers
@gerardo-w do not forget to mark the topic as SOLVED 😉
RE: Filamentum PLA Cristal Clear: Mid-Print no more layers
@antimix,
yes. For me the issue is theoretically solved. Unfortunately I got rid of the PLA that always jammed. In a certain way my statement that some PLA's are not compatible with a particular 3D FDM printer is not that wrong, since it is now clear that the heat break design has definitive an influence on the jamming for some filaments. I wonder now, if the general performance of the MMU2S will be even lower by replacing the PRUSA's original heat break with the unmodified E3D-V6-175-BREAK.
Since @volker-jaenisch started this thread he is the proper (better) one to mark the topic as SOLVED.
RE: Filamentum PLA Cristal Clear: Mid-Print no more layers
I wanted to post a pic showing the difference in the filament ends after changing the heat break. It is hard to tell in the pic, but the diameter of the filament end from the Prusa heat break (Purple) measures 2.19mm. The diameter of the filament end after switching to the E3D heat break (Red) measures 1.99mm. I have no idea why the Prusa heat break plug is longer and I would be interested in hearing everyone's thoughts about that. I assume it has something to do with the fact that there is no gap for the filament to squish into with the E3D version. So far (fingers crossed) I have had no jams or problems with filament loading/unloading since the change.
RE: Filamentum PLA Cristal Clear: Mid-Print no more layers
Dear @antimix!
Since this topic is started by myself I do not like it to be closed, yet. Today the new original E3D heatbreak arrives and I will change it ASAP.
Yesterday I found out that the PTFE-Tube in my MK3s hotend (yellow dot) ordered on 28.08.2019 as a kit is a non standard 4.15mm outer (2mm inner) diameter tube. I needed pliers to get it out of the 4.00mm hotend. I exchanged this tube by a 4,00mm outer (1.85mm inner) diameter one from the spare parts which look like the ones @gerardo-w recommends for the MMU2.
First test is encouraging, second test is still running.
But I am stil a bit puzzled about the tube diameters. From my opinion it does not make sense to have 1.85 mm PTFE-Tube bottelneck in the hotend independent of the heatbreak (2mm vs. 2.3mm) problem. I would think that a uniform 2.00mm diameter over the whole length of the hotend (tube and heatbreak) may be the best solution. Has anybody done this, before? I thought I read a forum post about it, but do not find it again.
Any help appreciated
Volker
RE: Filamentum PLA Cristal Clear: Mid-Print no more layers
Hi Volker,
I have the 2.3 PTFE tube connected to the original E3D heatbreak, not the 2.0mm.
This because the filament head, even with dribbling, sometime may be not very well formed, and may have a little ball shape. Having a slide larger PFTE tube, avoid the friction of the ball passing through the PTFE tube.
So, once extracted by the strong extruder motor, in order to minimize also the effort of the filament retraction by the MMU2 gear (that is not so strong, and it is inclide to shift) , I have replaced the original prusa tube from the extruder to MMU2 with a 4mm PTFE tube, that has an internal diameter of about 3mm. This is enough large that malformed shapes never jams inside it or create friction for the MMU2 gear. 😀
This ensure an always realiable filament extraction and insertion from MMU2.
Regards
RE: Filamentum PLA Cristal Clear: Mid-Print no more layers
Dear @antimix
Where to buy such (2.3mm) tubes?
Cheers,
Volker
RE: Filamentum PLA Cristal Clear: Mid-Print no more layers
Dear @all!
Yesterday I printed some Test-Parts utilizing PLA Crystal clear and the 1.85mm PTFE-Tube in the hotend. The 1.85mm Tube solved the PLA problem. I was able to print at lest the triple ammount of test-parts without a single jam! In comparison to the non standard 2.0/4.15mm tube that was delivered assembled in the MK3s extruder hotend. This made me very happy.
BUT
Then I printed PETG. After20 min of printing with the 1.85mm tube and the 2.00/2.30 heatbreak it stuck badly. This is my first in hotend-Jam that I was not capable in dragging out. Had to remove the PTFE Tube from the hotend. Also this is Prusament PETG. Had not a single JAM with PETG before and severals spools printed.
The bulge diameter is 2.20 mm. So the PRUSA recommendation cited in this thread to use the 1.85 tube for the MMU is to be questioned:
- I may be an advantage for MMU users only printing multicolor PLA.
- But for PETG it is a disadvantage and a pain to resolve.
I hand crafted a 2mm PTFE tube for the hotend. With this tube (and still the 2,00/2.30 heatbreak) I had zero problems PETG printing the same object.
Tomorrow I will change the heatbreak to E3D original. Are there any recommended manuals (beside the one from Prusa) for this operation available? What are the pitfalls?
Cheers,
Volker