Hi all. Again a quick question regarding extrusion.

(Lukas Obermann) #1

Hi all. Again a quick question regarding extrusion. When I extract 100mm if actually extracts 100mm. Every time. When then retracting 100mm it retracts 101-102mm. Every time… Any hint on where to look for this issue? I already decreased the microstepping, checked the hobbed bolt and the bowden. Using an Infill 3D Geared extruder.

Any hints welcome. Thanks.

(Neil Darlow) #2

When you talk of extraction and retraction, I assume you mean forward and reverse feeding of filament through an extruder body?

Is this an unconstrained length of filament i.e. not feeding from a spool?

I can imagine you would only get the same feed amount in both directions if the loading is the same in both. A significant load in one direction might lead to some slippage which is not evident in the other direction.

In practice this should not matter because you calibrate E_STEPS_PER_MM for forward feed and adjust retraction independently for best results.

(Lukas Obermann) #3

@Neil_Darlow yes, exactly. No Hotend connected. Just the extruder and the bowden. Is without a spool.

I can not see any difference in loading. Even when holding the filament, to provde some counter force, it still behaves the same way …

(Mateusz Perlak) #4

Are retraction speeds different? - did you tried to use same g-code for one and other directions.
If you use pronterface or other tool to retract and extrude - take a look at commands you send to printer - are they similar in values?

(Lukas Obermann) #5

@Mateusz_Perlak I actually sent it manually, “G1 E100 F300”, trying feedrates up to 600, consistent result…

(Mateusz Perlak) #6

weird, maybe measurement process is not correct… nothing else comes to my mind

(Øystein Krog) #7

When extruding it may compress the filament slightly, perhaps the compression translates to an elongation of the filament?

(Carlton Dodd) #8

I think you’re trying to fix something that seems well within tolerance.
You have a 1-2% difference, only on retraction. For a long retraction on a Bowden system, you might order 5mm and get 5.05mm. Even if you don’t have a purge after extract, you only end up 0.05mm short when you go to push filament again, which is insignificant. This is a high example. For direct-drive, the retraction is even less.
Am I missing something?

(Jim Peterson) #9

Is it multiplicative or additive error? Does retracting 10mm typically do 11-12mm, or 10.1-10.2mm?

(Lukas Obermann) #10

Øystein Krog, If correct shouldn’t it extrude more, instead of retract more? If the Filament get compressed and longer when extruding?

@Carlton_Dodd May be yes, I am just trying to fix my print quality, and I assume it has something to do with exstusion. Going to post a some new pictures soon. Maybe I am on the complete wrong track here.

@Jim_Peterson When goind 10mm it does around 10.1. Hard to meassure, but when repeting it it slowing retracts more then extrudes.

(Øystein Krog) #11

Hmm, do you have any backlash in the extruder?
Double check that grub screws are tight and that gears/belts are tight when moving back and forth.

(Jim Peterson) #12

Interesting. I’m surprised that is causing any printing issue. If you were to extrude and retract 100mm ten times each, alternating, would it end up 10-20mm behind where you started, or only 1-2mm? (In other words, is it cumulative, or just some form of initial settling?)

(Lukas Obermann) #13

@Oystein_Krog Can’t see any. Everything thight and filament is reacting instantly when turning the gears.

@Jim_Peterson Made some test gcode now where it extracts and retracts 100mm of filament 20 times. And, surprisingly, after 20 times it is still only 1 mm short. So seems to be initial. Thanks for the tip!