Issues with multiples and crashes  

RE: Issues with multiples and crashes

I would turn on 7x7 mesh bed levelling if it's not already on.

I would allow Magnet Compensation, 

And I would give the build plate a rigorous clean with hot water and dish soap. rinse with hot water and dry immdiately with plain paper towels

then I would use 'Life Adjust Z, My Way' and adjust the live Z

then I would try printing your models again. 

regards Joan
 

RE: Issues with multiples and crashes

So I've been running the 7x7 with mag compensation.  Between every print, I wash with hot water and dawn dish soap, I rinse and wipe with a paper towel.  I then after letting it dry, do a wipe with 97% isopropyl alcohol to make sure there are as few contaminates as possible.  I've had some success today switching back to the smooth build plate and dropping the z to the point where I have a smooth base first layer on the underside.  I ran some small parts today for some prototype pieces (I use 3d prints primarily to test things before I machine them) and it went okay.  I'm going to bite the bullet and try running the job again overnight.  The crash protection is a very nice feature, though I don't know why when you elect to not continue the print it moves to the middle of the build plate where it certainly will smash into whatever's there.  Moral of the story, if you're not going to continue printing, clear the build plate first then go through the prompts.

One thing I noticed is that my infill speed on the factory profile is basically double what I've run in my well dialed in Cura profiles.  I also generally use cubic infill as the structure model. Interestingly, the net print time is very close between the factory high speed .2mm profile and my high speed .24mm profile in cura when test slicing.  All of my crashes have been by products of catching on the infill.  Obviously, the thicker layer height is a factor, but I was surprised at how close the times were with the same infill percentages, the same files, and very different speeds on some parameters infill being one of them.

One thing I thought of today is that I have the Prusa setup in my office on a shaky desk as opposed to the incredibly stable and dense base the other machines are on.  I'm actually going to move it to a trashed granite surface plate as it's base later this week when I have some time to re-arrange things.  During the very fast, short distance moves, the shaking is violent.  That's also when I have the parts getting ripped from the bed.  I suspect in addition to contributing to some ringing and other print quality issues, that this may be contributing to the failures.

One thing I thought of was reducing infill acceleration (at least that's the setting in Cura I'm familiar with).  I suspect there must be a similar adjustment in Prusa Slicer, which I am very impressed with I might add.

Thoughts?

Posted by: @e4r104

Hi,

I've been printing for a few years, but only recently have gotten a Prusa I3 MK3.  I was fortunate to get a new, pre-assembled printer from someone who received it as a gift and never used it.  So it is brand spankin' new.  So far, I like it quite a bit and am impressed with the ease of use. 

I'm having an issue with some production PET-G parts.  This is an issue I've had in the past with my other unamed printers which took a lot of dialing in to resolve.  Basically, I have a number of smaller parts in a grid.  As the height gets to the 15mm - 20mm range, one often (not always) gets knocked off the build plate and causes a big crash.  I'm currently printing with Prusament PET-G using the .2mm layer height quality preset with the infill changed to 50% and a brim added with a .2mm separation.  I'm migrating to Prusa Slicer from Cura, so I'm still using pretty much canned presets as I've been told they work very well with the Prusament filament (this has proven thus far to be very true).  It is worth noting it is a new filament package and after having some issues, I dried it at 165 deg f for 8 hours to see if that would help.  I'm printing on a textured build plate.  Honestly, the bed adhesion seems quite good, it's just on these types of files I'm having trouble not having one piece get knocked off. 

While this is an issue on my other machines at times, I was hoping to avoid what I did to resolve it, which was to spend a lot of time with a dial indicator and brass shim stock to get the bed leveled so that the variance reported by my bl touch on a 4x4 grid was under .1mm across the board.  It's as miserable as it sounds.  The MK52 bed is sooo much nicer as is the new textured build plate.  I have a hard time seeing mechanical variance being the issue, that said, I haven't actually looked at what it's reporting as the variance as I haven't figured out how to do that yet :).  I may just need to slow things down, but I'd be appreciative of any other advice.

Also worth noting, I print using octoprint.  Given the firmware is based on Marlin, are the G and M codes more or less the same?  With my other machines, I've done the bed calibration tests using M commands from the console in Octoprint.  Can I do the same with the Prusa?

Link to print file

Link to print config ini

RE: Issues with multiples and crashes

use Expert mode, to access all of the settings
Regards Joan

RE: Issues with multiples and crashes

Thanks, I did find the accelerations doing that and I've changed them.  Still having the same issues sporadically.  I think I may need to do a bed level correction.  I'm going to make a mount for a dial indicator and get a sense of how flat the mk52 is.  Are there steps users have taken to correct issues there with shims or other means of getting it flatter?  

One thing I may be missed is that it suggests running the z calibration routine with pla.  I've been doing it with PETG (Prusament) and I noticed that the PET temp settings are lower than what it prints at. Is this intentional?  Should I only do the calibration with PLA and PLA heat settings?  I don't intend to use PLA on this machine.

Thanks again for the help.

Posted by: @joantabb

use Expert mode, to access all of the settings
Regards Joan

RE: Issues with multiples and crashes

Search Nylock Method, and Silicone method. 

One uses nylock nuts to crat a 'Jacking, action in place of 8 out of 9 heatbed spacers
the other does the same thing with short lengths of silicone tubing

regards Joan