Print failing at the exact same spot (layer 351)  

RE: Print failing at the exact same spot (layer 351)

on my mk4 i had lots of trouble with the rectengular infill on tall objects.

Maybe try an infill wich does not cross lines.  Like honeycomb or similar

RE:

Thanks, but I don't think that's the problem.

I had printed 7 with the same parameters, but sliced before the new firmware and slicer came out (but rotated by 45°). I now sliced it again in the same orientation as before, I also uploaded the old gcode so I can try multiple other things. I somehow still think that it's a slicer error.

Also, grid hits the other lines, rectangular does not.

RE:

My guess is it is this overhang is curling and causes a collision. If you can zip your project file (.3mf) and post it, that would help. And a close-up of the top of the failed print - especially overhangs like the one I've circled.

And you mean rectilinear infill, not rectangular, right? Rectilinear alternates infill direction by layer and does not cross. 

RE: Print failing at the exact same spot (layer 351)

It's not the overhang. I printed 7 of those without any problems and there are hundreds of those Cores in circulation.

I did print it again rotated by 25° and it worked without a problem. 

So I guess this leaves me with two options: It was sliced incorrectly and I printed that twice. After I resliced it I

1. either deleted the new file or it was sliced incorrectly again in the default orientation or
2. the Slicer caches files that it has sliced before so it does not have to recalculate it. So rotating it forced it into a completely new calculation.

I will never know what it was because now I have a working file again. Thanks for all your input.

RE: Print failing at the exact same spot (layer 351)

No caching whatsoever.  Overhangs like that are treated differently in Slicer, thus no two are the same, and a new slice may very well have not caught it and printed it fast rather than slowing down while printing it. 

If it's working again, great; but keep in mind that type of pointy overhang is problematic and they do tend to curl upwards.

RE: Print failing at the exact same spot (layer 351)

It's a bug, I've been aruging with Prusa about it for 2months, I  raised layer shifting as a serious issue back in July. 

I gave them a file, they reproduced the fault, accepted it was gcode generation / firmware and emailed me and nothing has happened since.  I've tested the new alpha firmware, it's not fixed.

Other than mechanical issues (you need to rule these out first like belt tenstion etc.), there are settings that bring this one and can be reproduced consistently.

1) "avoid crossing perimeters" when this setting is on I get layer shifting all the time with the CORE ONE.

2) Organic supports, I consistently get layer shifting at the same spot like you for the same print. I've now got a broken fan shroud as a result of the collision. I'm not testing other supports.

I have 3 CORE ONEs and can reproduce this bug on every one. This does not happen on any of my 20+ MK4 printers or my XLs. I've almost given up speaking to their support as they don't listen and have to resort to shaming them here now. I do a lot of commerical prining now and I need to move away from Prusa as this is too sloppy and costing me a lot of time, money and materials. 2+ months they've had to fix this.

RE: Print failing at the exact same spot (layer 351)

Yeah it's a pretty well known bug now. The latest alpha firmware was supposed to have several fixes for layer shifting, its being reported as still not fixed.
I have one print where it happens a lot, and I have to print a couple of these parts over the next couple of days and I'm dreading it. 
The machine makes a really weird noise when it happens, almost a screech like the motors are being driven too fast leading to missed steps. The only way I've found round it is to slow the print right down....but that's not really a solution, it's just avoiding the problem.

RE: Print failing at the exact same spot (layer 351)

The problem is they are unwilling to accept it's a bug, it's infuriating, and even after they sent me an email confirming it was originally. I chased this yesterday and was told:
------------------------------------------------------------------
"If you're having issues with other parts, then the problem is mechanical.

The external perimeters first option is not compatible with Arachne perimeter generation. We're currently not working on a fix for this, as it was never intended to be compatible. If you require printing external perimeters first, please switch to the standard generation.

As for any other cases, the issue is purely mechanical, so in order to effectively troubleshoot the problem, please share a video of the printer homing with me and revisit the belt tensioning step of the assembly."
-------------------------------------------------------------

The problem is not external perimeters, it's "avoid crossing perimeters" and other several other settings that seem to trigger the issue. I believe it to be when the print head has to make a lot of moves in a single layer (which is increased with "avoid corssing perimeters"), it crashes and layer shifts. This also happesn when using suppports for more complex parts where there are a lot of print head moves. This bug is not on the MK4 or XL (I can print the same parts perfectly), my problem is I need the heated chamber for PA prints so this has really messed me up. I'm now in the process of making make-shift cases for my MK4 printers to work around this and I'll be returning several of my CORE ONEs as they are not fit for purpose.

I have spoken to some arrogant support employees at Prusa who insist bug is simply "impossible". We need someone in Prusa to take responsibility for this as the first line support are troublesome barrier, their lack of understanding is preventing this getting fixed.

RE: Print failing at the exact same spot (layer 351)

Prusa are spending loads of development time making belt tensioning easier, strobing effects etc. probably becuase so many people are contacting them about layer shifting and they're assuming it's mechanical, when it's probably not in most cases, it's the issue highlighted in this post. I wish that had spent the time fixing this bug....

RE: Print failing at the exact same spot (layer 351)

This is where I'm at, CORE ONE not fit for purpose, so i've had to resort to hardboard cabinets and polythene in order to produce more complex PA parts. The MK4 does not exhibit this layer shifting bug. How about fixing it Prusa?

RE: Print failing at the exact same spot (layer 351)

You would think with their experience with the XL that they could have just "scaled" it down when creating the Core One. So, I guess there was a rush to get it out the door so they don't fall further behind. Which makes us all beta testers. With regard to shaming Prusa on this forum, the indications are (to me at least) that no one from Prusa hangs out here. 

It's funny that this is first Prusa I've owned where I've had to preemptively print spare parts (like the fan shroud) in an attempt to mitigate future disaster. Fortunately I'm just a hobbyist/tinkerer so downtime doesn't affect me in the same way. If I hadn't of sunk so much time and money into it I probably would have taken a sledgehammer to it by now. I have my eye on the Snapmaker U1. Six months after it's released, if the there are no major issues, I'm giving the C1 to my niece and buying that as my replacement. 

RE:

To really bring eyes to an issue this is the worst place for it, there’s currently 12 users online 😂

Do it on Reddit, tag the new moderators who from my understanding have the actual role of being a conduit between the community and Prusa the company.

RE: Print failing at the exact same spot (layer 351)

It would have been cool if they came out with a Prusa XL Mini (yes, that cancels out so maybe the Prusa Medium?) A smaller XL with a 2 head changer. Fingers crossed that the INDX is not vapourware.