M600 is broken
Hello. I finally managed to get a "successful" dual multilayer, multimaterial, multicolor print with stock MK3S and manual filament swap. Essentially what MMU does but manually. I started 3d printing personally 6 months ago (even though i had some experience before) and it amazes me this is not a feature in any of the slicers available.
I tried using PrusaSli3er, Cura and Simplify3D. Cura is a no go, Sli3er using a modified profile from Rainer ( http://schlosshan.eu/blog/2019/03/02/prusa-i3-mk3-real-multicolour-prints-without-mmu/ ) kinda works but i cannot get the detail i need. S3D needs some manual gcode work but here is the result with 0.25mm nozzle, white Prusament PLA, and black Prusament PETG:
https://imgur.com/gallery/XTL0Sbk
Other that polishing i have to do with the gcode there is one main problem:
M600 command is "broken":
Not only after confirming filament change it will extrude while moving to the part, but the first extrusion makes a blob possibly by an internal G92 E0 command?
I did some research and found this:
https://github.com/prusa3d/Prusa-Firmware/issues/197
https://github.com/prusa3d/Prusa-Firmware/issues/904
The first unwanted extrusion seems to be intentional but it is at best stupid. Since you extrude to flush the previous colour there shouldn't be the need to do this.
Someone said is to equalize the pressure? Maybe it overpressurises that's why i get a blob after? I have no idea. Here's my gcode:
https://www.dropbox.com/s/bdpe7m36lnh75im/key-1.75u-dual-shot-test-7.gcode?dl=0
Go to line 998. I get a huge blob at line 1019/1020.
Can we get this fixed?
Kind regards
RE: M600 is broken
This is a peer-to-peer forum. Bugs need to be reported on the appropriate GitHub pages.
RE: M600 is broken
This "issue" is opened on github in the 2 links shown in my post, nothing was done because they claim is a "feature"
RE: M600 is broken
I may be barking up the wrong tree, but have you tuned your K factor (Linear advance), this affects nozzle pressure between speed changes, and may (I don't know exactly) also have an affect when starting to lay down filament as technically there would be a speed change.
http://marlinfw.org/docs/features/lin_advance.html
This needs to be tested for each filament.
Normal people believe that if it ain’t broke, don’t fix it. Engineers believe that if it ain’t broke, it doesn’t have enough features yet.
RE: M600 is broken
@chocki
I haven't touched this and never had any issue if my program doesn't use M600. It is obvious the extruder is actively extruding. The only part i cannot say for sure is M600 related is the first extrude after M600 is done like i mentioned in my post.
RE: M600 is broken
This is a problem for me too.
I haven't looked at the gcode, but I do experience the unwanted extrusion after a color change.
I just came here to check if there was some slicer setting I needed.
For now, I just let the final color change plastic harden. Then I hit the button, and the unwanted plastic is hopefully still connected. So then I can remove both pieces before the new extrusion starts again.
I subscribe. Hopefully there will be a true fix to this issue.
RE: M600 is broken
This is a problem for me too.
I haven't looked at the gcode, but I do experience the unwanted extrusion after a color change.
I just came here to check if there was some slicer setting I needed.
For now, I just let the final color change plastic harden. Then I hit the button, and the unwanted plastic is hopefully still connected. So then I can remove both pieces before the new extrusion starts again.
I subscribe. Hopefully there will be a true fix to this issue.
I haven't looked it up in more detail since then but it is definitely an issue with the command in the firmware. I guess you can re-compile the firmware with changes so it doesn't do this. They didn't even look at it.
RE: M600 is broken
@kyriakos-g
I've done a number of multi-color prints (typically changing at a specific layer) using M600 and haven't had any issues. You have to hold on to the extruded filament before hitting "Yes" so that you pull off the extra extrusion, but other than that it's been working well for me.
RE: M600 is broken
@kyriakos-g
Technically it does make sense because during the waiting time some filament is drooping out of the nozzle but you want to start the print with fully loaded nozzle. I‘m also holding the filament while the head moves. This way you have always a clean start.
Just removing the last extrusion code will reduce the quality of the print, so it can‘t be a right solution. What‘s your proposed solution for this?
Often linked posts:
Going small with MMU2
Real Multi Material
My prints on Instagram
RE: M600 is broken
@mdaneman
There shouldn't be the need for extra work. Problem is that the command is extruding after clicking yes. This shouldn't happen in any logical way.
@nikolai-r
Did you even see the issues i linked on github? It is the same issue. You always use a wipe tower and that's the proficient solution. The firmware shouldn't do stupid sh*t like this. The other solution is to give an option to enable or disable this behavior. Or have an alternative command if you don't want to touch this one. LOTS of solutions. It might beneficial to some, but the machine extruding when it's not given an extrude command is BAD ENGINEERING 101.
RE: M600 is broken
@kyriakos-g
No need for a rant. M600 is not using any wipe tower as it would be waste of material. It's extruding in the air and don't have any base for a clean purge like during regular MMU print.
If you have an idea for an improvement, then try to describe it briefly and place a feature request on GitHub. The two linked GitHub issues were clearly answered by the devs as not being an issue. If you want to change it by your own, feel free to do so and post your results.
Often linked posts:
Going small with MMU2
Real Multi Material
My prints on Instagram
RE: M600 is broken
@kyriakos-g
No need for a rant. M600 is not using any wipe tower as it would be waste of material. It's extruding in the air and don't have any base for a clean purge like during regular MMU print.
If you have an idea for an improvement, then try to describe it briefly and place a feature request on GitHub. The two linked GitHub issues were clearly answered by the devs as not being an issue. If you want to change it by your own, feel free to do so and post your results.
Did you even watch my video in the first post? It is DEFINETELY an ISSUE. That's why the issues are still open in github. Are you some kind of politician that denies the problems in face of evidence?
RE: M600 is broken
@kyriakos-g
No need for a rant. M600 is not using any wipe tower as it would be waste of material. It's extruding in the air and don't have any base for a clean purge like during regular MMU print.
If you have an idea for an improvement, then try to describe it briefly and place a feature request on GitHub. The two linked GitHub issues were clearly answered by the devs as not being an issue. If you want to change it by your own, feel free to do so and post your results.
Did you even watch my video in the first post? It is DEFINITELY an ISSUE. That's why the issues are still open in github. Are you some kind of politician that denies the problems in face of evidence?
RE: M600 is broken
An ooze shield might work as a quick workaround for your problem it would catch the bulk of that tail before it made it into the print area again.
RE: M600 is broken
An ooze shield might work as a quick workaround for your problem it would catch the bulk of that tail before it made it into the print area again.
No that will not work because it is lots of material. I will compile the firmware myself and modify m600 since they don't even care.
RE: M600 is broken
I think they've looked at it and disagreed. Opening an issue on GitHub is no guarantee that your suggestion will be incorporated into the main builds, nor even addressed. The official stance seems to be that guaranteeing an even flow of filament is the priority. This may be from a simple support perspective. Fortunately, that doesn't mean you're stuck.
One approach would be to use Octoprint as essentially a pre-processor, intercepting the gcode as it's sent and substituting your preferred sequence. You might not issue M600 but substitute a script that accomplishes the same tasks. This would have the advantage of working with any slicer. There's a write-up by somebody doing other M600 massaging here that might get you started.
Marlin does seem to have ACTION_ON_PAUSE and ACTION_ON_RESUME hooks that you can enable (uncomment) if you're OK with compiling your own firmware. I have no idea if those are present in Prusa's firmware fork.
I've just started on the Octoprint plugin path for other issues myself, but unfortunately work kept me busy most of last year. Hopefully these hints will help you get started. Please report back with your solution.
RE: M600 is broken
Good luck, kyriakos-g.
I just assumed it was a PrusaSlicer issue.
And I have no plans of editing gcode, nor of using Octoprint as an alternate fix.
They are so proud of the color change feature, but it pukes extra filament every color change. Maybe they are trying to trigger more MMU2 sales.
Bigger worries right now, as struggling with first layer issues.
RE: M600 is broken
[...] They are so proud of the color change feature, but it pukes extra filament every color change. Maybe they are trying to trigger more MMU2 sales.
I just snag the extra filament with tweezers as it moves into position. I have to be there for the swap anyways. Doesn't this work for you?
RE: M600 is broken
[...] They are so proud of the color change feature, but it pukes extra filament every color change. Maybe they are trying to trigger more MMU2 sales.
I just snag the extra filament with tweezers as it moves into position. I have to be there for the swap anyways. Doesn't this work for you?
I already said that I did.
Upgraded my firmware, re-calibrated, and now have very good first layer. Haven't tried a color change again yet.
RE: M600 is broken
So this is literally a complaint about having to snatch a second stray bit of filament as the nozzle returns to printing? I can see why it's not a priority.