2 colour sheep example taking almost 7 hours! Is that right?
Hi,
I've just assembled my MMU2S upgrade on MK3 and am trying my first 2-colour print, using the example sheep, downloaded from the Prusa site.
The gcode from Prusa indicates: estimated printing time (normal mode) = 2h 43m 16s
The gcode from Slic3r (Prusa version) indicates: estimated printing time (normal mode) = 6h 45m 50s
I loaded the two .stl files and selected wipe into this object's infill. Printing layer height is .2mm, using Prusa's defaults, except for changing fill to 10%, using the rectilinear fill pattern.
What I can't understand is why there is such a massive difference in print-times for the gcode form Prusa, compared with the code generated by Slic3r cv 1.4.3 (win 64). Can anyone suggest why, or recommend things I should check.
FYI: Filament changes are taking around 35 seconds from retraction to next head movement. Is that about right?
Thanks for any advice.
Best Answer by parkis:
The difference between print times is because in g-code provided by Prusa is sheep scaled down to 50%.
RE: 2 colour sheep example taking almost 7 hours! Is that right?
2h 45m doesn't sound right at all for a print as big as the sheep. My guess is that's a single color gcode file.
Try doing a single color slice yourself and see if that is close to their gcode time.
RE: 2 colour sheep example taking almost 7 hours! Is that right?
Hi, @gnat - thanks. That's a good point. Normally, if I'm doing something for the first time, I automatically assume it's me that's got it wrong - maybe not this time, though.
The sheep almost printed properly, but the MMU failed a filament change about 30 mins before completion. No matter what I did, it wouldn't recover - the idler and selector seemed to be having a disagreement about which filament to load - the selector wanted filament 1, but the idler was on filament 2 🙄 Even an MMU-only reset would not clear the issue. And now the extruder isn't extruding properly. So, I plan to take it apart and check for issues, this evening, before trying again.
Once I've go things working again, I'll try your suggestion
RE: 2 colour sheep example taking almost 7 hours! Is that right?
roughly 90 seconds per color per layer.
I have a Prusa,therefore I research.
RE: 2 colour sheep example taking almost 7 hours! Is that right?
The difference between print times is because in g-code provided by Prusa is sheep scaled down to 50%.
RE: 2 colour sheep example taking almost 7 hours! Is that right?
Thanks, everyone, for your advice.
Indeed, yes, the downloadable gcode is much smaller than that generated if you slice the .stl files, wiht no scaling.
I printed the 100% model in around 6:45 (no supports) and the 50% model (with supports) in around 2:40
For anyone else who comes up with the same question I did, I've attached a pic to show the size difference:
RE: 2 colour sheep example taking almost 7 hours! Is that right?
Hi, @gnat - thanks. That's a good point. Normally, if I'm doing something for the first time, I automatically assume it's me that's got it wrong - maybe not this time, though.
The sheep almost printed properly, but the MMU failed a filament change about 30 mins before completion. No matter what I did, it wouldn't recover - the idler and selector seemed to be having a disagreement about which filament to load - the selector wanted filament 1, but the idler was on filament 2 🙄 Even an MMU-only reset would not clear the issue. And now the extruder isn't extruding properly. So, I plan to take it apart and check for issues, this evening, before trying again.
Once I've go things working again, I'll try your suggestion
What did you do to fix your issues? I am also having the problem where the idler and selector are not synced up.
RE: 2 colour sheep example taking almost 7 hours! Is that right?
Hi @justin-w17
It turns out I had a few issues (perhaps one led to the others).
Before I start, if you haven't done so already, make sure you have "calibrated" the extruder's IR filament sensor - with out this setup correctly, things will just keep going wrong! The steps for this are in the user manual - it's just a question of properly aligning the "chimmeny" on top of the extruder body (so the lever/stick on the idler plate interrupts the IR sensor at the top of the "chimmeny").
OK, my issues: I explain the causes of these as I talk about each one but please remember, I wasn't aware of the cause of each, at the time I was running into problems...
1. At some point I must have knocked the connector to the extruder's IR sensor - this meant the controller couldn't tell when filament was in the extruder. This may have caused one or more of the other issues (as the MMU would have been trying to push filament, even though it had already reached the extruder
2. I think the knock on effect of item 1 was perhaps the cause of this (bear in mind, I didn't know this at this time). I found filament wouldn't even properly load/unload from the MMU2S. The motor kept turning and got hot. I tried adjusting the tension on the 2 screws/springs for the idler, but to no avail. On looking inside, I could see everything was as it should be, except the filament wasn't moving. On closer inspection, it turned out the filament had simply been ground away by the gears (dust in the unit, with a U-shaped cut into the filament), so there was nothing left to 'bite' on and push filament through. I scrapped the damaged filament and started over.
3. When the filament reached the extruder, it wasn't being pulled in by the bondtech gears even though the extruder motor was running. This turned out to be due to the gear having slipped down the motor shaft, inside the extruder body. It's difficult to see this has happened, unless you remove the idler plate and gear from the side of the extruder (I only found out much later - following a lot of misery). Now, I've never had that happen before (before I built the upgraded extruder for the S version) and I'm sure (fairly!) I tightened the grub screw properly when I first assembled it. So, either the messing about loading/unloading filament when the other issues were active caused this to become lose OR I just didn't tighten it properly in the first place.
4. One other issue I had, which I didn't mention previously: on my first print, the one that failed 30 min before completion, I noticed little beads of plastic had fallen onto the bed and even into the infill. It was only when I was fixing the bondtech gear issue, I noticed filament had been oozing from the hotend. I must have loosened that when disassembling/reassembling. Another lesson learned!
5. With so much going on, I never really got to the bottom of the selector vs idler "argument". What I think may have happened, with all the filament jamming and detection-failure which was going on, it's possible a piece of filament broke off and got stuck in the slector, preventing it from moving correctly. That said, I would have thought that condition should have simply stopped the idler and raised an error. However, I do know it's very easy to upset the MMU - using the LCD menu, load filament 1 to the nozzle, then tell it to load filament 2: it doesn't seem able to handle the fact there's already filament loaded and just throws an error (what I'd like it to do, is simply unload filament 1, first - simples!)
I'd be interested to know if/how you resolve your own issue. Please post an update when you can. Thanks.
Anyway, this is all fixed, now and thing are printing properly! And I haven't seen the selector vs idler issue since!
My only remaining issue is the MMU2S filament buffer doesn't seem to work reliably for me. I find unloaded filament sometimes pushes filament all the way back to the spool, in creasing the chances of it tangling and wrecking a print. It's probably a case of experimenting to find the optimum layout. However, as I want to adapt the whole setup to fit in my Lack enclosure (HERE), I'm searching for an alternative to the Prusa buffer in order to save space.
RE: 2 colour sheep example taking almost 7 hours! Is that right?
I finally had some time to look at this again and found my issue. The black and blue wires on the Extruder Motor on the MMU2s were reversed. So it was spinning the idler in the wrong direction.