Print not "printing" at seam of box, leaving a gap in the print.  

RE: Print not "printing" at seam of box, leaving a gap in the print.

Let me guess, you are printing multiple boxes at the same time ?  Its most likely filament leaking from the 0.6 nozzle during travels between islands.  

There are some threads about this effect both on the forum here and also a github issue regarding it.  There seem to be many contributing factors, like filament linear advance settings etc.  It also seems to be more prevalent on the mini with its bowden setup but it does appear on other printers.  The jury is out on if its a slicer issue or a combination of things.  The github issue if you want to read it and try any of the suggestions is located here https://github.com/prusa3d/PrusaSlicer/issues/11914

I'd start with experimenting with your retraction settings for the 0.6 nozzle.  With the Mk3 don't go more than 2mm.

RE: Print not "printing" at seam of box, leaving a gap in the print.

The problem also occurs when printing only one box. I believe that the printing of the perimeter starts and finishes at this point. Here is another  I just printed, and a .3mf project saved from prusaslicer. Currently the retraction length is just 0.8mm. Do you think I should set it to something like 0.2mm? I don't mind about stringing, this is a functional print. 

RE: Print not "printing" at seam of box, leaving a gap in the print.

A question, are you seam setting aligned or random?  

RE: Print not "printing" at seam of box, leaving a gap in the print.

No project made it through.  You have to zip files like that to attach them here.  

RE: Print not "printing" at seam of box, leaving a gap in the print.

Seam is aligned. Thanks for the tip about files, here is a 3MF and gcode, what do you think

RE: Print not "printing" at seam of box, leaving a gap in the print.

Thanks for attaching the project file.  Looking at the slice preview and turning on show travels I think your issue is a couple of things,  that wall is a single perimeter, the gap is at your layer start point.  However to get to that start point its travelling diagonally across from the furthest corner so in effect a long travel resulting in the missing filament on the next layer start. Single perimeter wall with travels are sometimes (often) problematic.  If you aren't already doing so, start making use of the slice preview and turning on some of the extra visualisation options like show seams and travel.  You can rapidly spot where the problem areas lines up.  It's also good practice and can save you a lot of time and filament in wasted prints in general 🙂

I'd thicken up that single perimeter wall just on general principles and re-slice to see what effect that has on the path planning. 
I played around with the seam position setting but even with Nearest it still has a long travel to get back to the start point due to planning around that single wall perimeter so I don't think that will help in this case.

I wouldn't normally recommend this but you could also experiment with Printer Settings>Extruder>Retraction>Deretraction extra length.  That will push out a bit of extra filament after the retraction.  Something like 0.1-0.2mm.  The problem is that can cause bulges/too much material on other areas.  So its one of those problematic settings. 

I suspect that making that wall a bit thicker to give 2 perimeters will be the best route.

RE: Print not "printing" at seam of box, leaving a gap in the print.

I tried printing with a random seam and there was no improvement at all but stringing did get worse.

Interestingly, the problem isn't occurring where all the deretractions are occurring but in another corner, highlighted in this photo.

I have attached the GCode and 3MF file.

After this I tried increasing the wall thickness from 0.68mm to 1.2mm, which of course means two perimeters are required. This has solved the problem and increased filament use from 20.6 g to 21.6 g, increasing print time from 50m to 51m. So, thanks for making me try that, I guess I had just pushed the optimisation too far without thinking.

I greatly appreciate your taking the time to teach me and share your experience here.

RE:

In case anyone else has this problem, I still get it on my 0.6mm nozzle with a 1.0mm thick wall (done in 2 layers). From examining the extruder path, it appears to happen at the start of each layer, not the end as I previously thought.

I think it is the same problem that others are having with benchy prints here and general prints here.

RE: Print not "printing" at seam of box, leaving a gap in the print.

Yeah, its not fully primed with material when its gets to the start of your new layer.  That's what the previously mentioned extra length on restart was for.  The bigger the nozzle the more this happens as bigger nozzles are generally 'leakier'.  

RE:

Posted by: @neophyl

Yeah, its not fully primed with material when its gets to the start of your new layer.  That's what the previously mentioned extra length on restart was for.  The bigger the nozzle the more this happens as bigger nozzles are generally 'leakier'.  

I didn't realise `extra length on restart` would apply to me as I only have a single tool (MK3S+) and the description says "When the extruder is compensated after changing tool, the extruder will push this amount of additional filament." I was also reticent because I saw 2mm+ of gaps on each layer and the other user said that 0.1mm `extra length on restart` ruined their model. 

However, I have tried 0.05mm extra length on restart and it helped, 0.1mm seems to have fixed the problem completely. I'll update if I have any more problems but how can I suggest that the description for this variable is updated? I suggest changing the tooltip to:

"When the retraction is compensated after changing tool or changing layer, the extruder will push "
"this additional amount of filament. This parameter is useful for closing up gaps on seams."
"Typical values around 0.05mm"

RE: Print not "printing" at seam of box, leaving a gap in the print.

Sorry, ignore me 🙂  The setting that is now called 'Deretraction Extra Length' used to be called Extra Length On restart.  Funnily enough it STILL is called that in Super Slicer and Orca Slicer so its easy to get confused if you use multiple slicers like me.  Not sure why the renamed it in PS

In fact if its helping you by setting it I too am surprised as the tooltip would indicate its applicable to multi extruder setups.  

RE: Print not "printing" at seam of box, leaving a gap in the print.

Posted by: @neophyl

Sorry, ignore me 🙂  The setting that is now called 'Deretraction Extra Length' used to be called Extra Length On restart.  Funnily enough it STILL is called that in Super Slicer and Orca Slicer so its easy to get confused if you use multiple slicers like me.  Not sure why the renamed it in PS

In fact if its helping you by setting it I too am surprised as the tooltip would indicate its applicable to multi extruder setups.  

Indeed, I think it was just a fluke from other settings I changed (sequential print and faster perimeter speeds), because the problem is back even with "Extra length on restart" at 0.2mm. 

I guess I just don't understand how it only seems to be happening on these sharp corners with "narrow" walls. There are some 3mm posts (cylindrical) in this model which have no apparent problems on the seams. 

I made a little test STL to iterate faster. Deretraction extra length values of 0.5mm and 1.0mm haven't made any difference to the problem at all. 

(De)Retraction speed had no effect. 

Seam gap distance set to -5% (instead of +15% default) also had no impact whatsoever.

RE:

Is this a bug in prusaslicer?

I'm still trying to get to the bottom of this. Is it possible that my slicer is ignoring my "deretraction extra length"? From the attached GCode file, with retraction set to 0.3mm and deretraction set to 0.1mm or even an outrageous 5mm, the deretraction is still exactly 0.3mm. I don't have any retraction overrides set in the filament overrides section, but they don't change anything either.

Excuse the poor photo, but here is the results with my notes.

If I disable "wipe while retracting" then the deretraction extra length is honored. I will try this next (without deretraction extra length)

RE:

So, disabling "wipe while retracting" has solved the problem completely. 

In the GCode for my previous post, I noticed that the feed rate during the wipe is 8640mm/min or 144mm/s, which is a LOT faster than the alleged retraction speed:

Could a bug in the implementation of wiping be responsible for this , with the extremely fast retraction speed (and blocking of de-retraction extra length) being the root cause?

RE:

So, after a lot of getting my head around  the GCode, my post above was wrong. The feed rate is set to 144mm/s but because the amount of extrusion requested from the extruder in each GCode movement is very low, the time taken to move that distance (even at 144mm/s) ensures that the filament isn't extruded faster than the set limit (default 35mm/s).

However, this seems to be too fast for me. By setting it to 6mm/s I got much better results, and this is observed even when wipe is enabled.

So there is a small bug in prusaslicer (ignoring the deretraction distance when wipe is enabled) but it wasn't the root cause of the problem.