Filament Extruded after Filament-change bevor Wiping on Wipe-Tower
 
Notifiche
Cancella tutti

Filament Extruded after Filament-change bevor Wiping on Wipe-Tower  

  RSS
sergio.s2
(@sergio-s2)
Active Member
Filament Extruded after Filament-change bevor Wiping on Wipe-Tower

Hi There,

I am printing the fantastic "Split-Flap-Clock" recently posted on the Prusa Homepage. I have a MK3S with no MMU, so I need to change the Filament myself. No big Deal, Working fine so far.

There is one issue: I use a Wipe-Tower, to be sure the first time the printer starts to print the acutal Element (Flap for the Clock in my case) it's perfectly ready with the right color. When I have changed the filament, I remove the Filament the printer pushed out while changing and I confirm that the Filament comes out with the right color. I Expect the Printer than to go to the Wipe-Tower and print one layer with the new color (which sometimes has a bit of the wrong color still left, but thats what the whole Wipe-Tower is for!) But just before it starts to Print to the Wipe-Tower it pushes out quite a lot of filament with no movement and then starts to print to the Wipetower, smudging the just (use-less in my case) pushed out Filament all over the place while I have to catch it, somehow! 🤬 🤣 

I use this Version of Slicer: 2.3.3+X64

And i Think the Problem is this section in the G-Code:

 

; printing object flap14.stl id:13 copy 0
;WIPE_START
G1 F8640.000
G1 X188.291 Y10.000 E-0.76000
;WIPE_END

Why does it do this G1 F8640.000 before it starts moving?

And where in the Slicer can this be disabled? Or is this a bug? (Besides the usual "bug" sitting in front of the Keyboard 😆 )

Thanks a lot!

 

Sergio

Postato : 10/08/2021 11:43 am
Neophyl
(@neophyl)
Illustrious Member
RE: Filament Extruded after Filament-change bevor Wiping on Wipe-Tower

That gcode is not the problem.  In fact that code is telling it to retract filament by 0.76mm.

What you are coming across is built into the mk3's firmware and can not be altered without changing that firmware.  Its also design intent apparently.  The 'correct' process for a filament change is to not remove the filament but to hold it with a pair of tweezers and then when its starts moving and extrudes that extra length it will pull away from the extruder and be left in the tweezers and not on your print.  Its basically priming the extruder again.  Thats the theory anyway.  Its been reported numerous times over on github.  For example https://github.com/prusa3d/Prusa-Firmware/issues/2502 or https://github.com/prusa3d/Prusa-Firmware/issues/197  

Postato : 10/08/2021 12:12 pm
sergio.s2
(@sergio-s2)
Active Member
Topic starter answered:
It's not a bug, its a feature!

Hi Neophyl!

Thanks for your quick and precise answer!

So we have a classical: "It's not a bug, its a feature!" situation 🤔 

Well, I don't get it at the moment (so it seems do other users, thus the github issue reports) but as it's built in the firmware I will do what I already did: Grab the pushed out filament, move it slightly away (now there is just a thin cobweb connection to the nozzle) and then pull it away as soon as it has stopped sh**ing in the air again and starts printing to the wipe-tower. 🤪 

 

Best Regards!

Sergio

Postato : 10/08/2021 4:37 pm
Neophyl
(@neophyl)
Illustrious Member

Yep 🙂  and note that the earlier link I posted is issue 197 so really really old. 

There's a bunch of posts here on the forum too somewhere but finding anything on the forum is not easy.  It was easier to find the github issues as at least the search on there works better lol.

Postato : 10/08/2021 7:43 pm
sergio.s2
(@sergio-s2)
Active Member
Topic starter answered:
Different handling in case of planned filament-change?

I fully understand why they have to do it when you run out of filament mid-print. Sure you want to have the nozzle filled, pressured and ready to go, as the print just continues the very millimeter it stopped and it might be in a perimeter.

But!

Couldn't there be an other command than the M600 used when the user does a filament-change on purpose? Especially (as in my case) when I even decide to use a wipe-tower? But since this is done in Firmware, might not be so easy. M600 might also been used by other manufacturers. Seems to be some sort of "dead-end" where everybody is right. I understand the developpers point of view, still I am not fully happy with the situation as it is for me at the moment.

Perhaps this is more of an Enhancement Type of thing for github, than a bug. Perhaps the Slicer should drop something like a "M601" in case of a planned filament-change, especially when it will start with a wipe-tower section and in this case it should stabilize the pressure in the nozzle while "wiping the tower" and not "extrude in the air, before start". But I don't know how difficult it is to set up a new G-Code. Guess there is some sort of committee who needs to agree to it, and if it's just something for Prusa, they might refuse it.

 

Best Regards!

Sergio

Postato : 10/08/2021 11:21 pm
Neophyl
(@neophyl)
Illustrious Member
RE: Filament Extruded after Filament-change bevor Wiping on Wipe-Tower

As I said, its not a slicer issue so the gcode command is correct.  Its a firmware issue.  Theres a thread in the general forum asking what people think should be in the limited space left for the 8 bit firmware (ie Mk3).  I suspect that the same internal function is called for both, probably a decision made due to the limited space available in the hardware.  I think its also used for filament load/unload too in general use.  Might be worth adding that into the pot on that thread as a worthwhile addition/change.  

A request on github probably wouldn't hurt either.

btw adding a new gcode command would also need to be added to the firmware taking up that precious space 🙂  The actual M600 command definition for Marlin has several parameters that can be appended to the actual command.  However I don't know how many (or if any) are actually implemented in the Mk3 firmware to be used.  Its quite normal for manufacturers to 'turn off' commands that are not applicable to their device. 

Postato : 11/08/2021 6:28 am
Condividi: