Announcement

Collapse
No announcement yet.

Prusa Mk3 Bed Adhesion Problems

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Prusa Mk3 Bed Adhesion Problems

    Seems we are having problems with prints sticking to the bed of the new Mk3 printer.

    The first sets of prints worked awesome.
    On Monday night (Jan 28) the first layer wasn't sticking and was peeling off in chunks that can best be described as "scales".

    The following fixes were tried, and eventually worked:
    1. z-axis calibration
    2. first layer calibration seemed ok. (It was hard to tell if the filament was flattened enough with the fine layer)
    3. first test print had the same issues. A live adjust z-axis was performed: Initial z-axis setting was -0.48. I tried going lower and going higher, settling on going much lower at -0.70.
    This initial fix resulted in good 1st layer adhesion.

    It seems the problem has popped back up on Tuesday night (Jan 29), and us maker's need to figure out why.
    1. What has changed between prints?
    2. Is the live adjust z still set at -0.7?
    3. If we are constantly having to mess with live adjust z, there is an issue with calibration.
    The best way to get to the bottom of this is to track the issues as a community so we can get to the root cause.


  • #2
    How consistent is the removal and replacement of the bed? We are dealing with an error in fractions of a millimeter. Perhaps we should only remove the bed as necessary and be prepared to recalibrate if you do.

    Comment


    • Garret H
      Garret H commented
      Editing a comment
      The Prusa runs through a level check before every print that is supposed to account for that., although I could be mistaken.
      Because I did a live adjust mid print, I am not sure if that setting stayed in the system.
      If it did, and the printer does a level check every time, the issue could be in how the Prusa is measuring the distance to the bed.

  • #3
    Seems some of these bed adhesion problems can be chalked up the the G-code I've been running.
    Tried out a print tonight using the default Mk3 "0.2mm fast" setting, it worked well, but is anything but fast.

    Trying out some new Gcode now with a different first layer setting from the last batch that had problems.

    Comment


    • #4
      Did you have success with the GCode changes? What settings did you use?

      I'd like to try the MK3 for some miniature models....

      Darren

      Comment


      • #5
        Yes, two different instances with great results.

        I took a well proven STL (Vaughn's stuff) and printed it with the Prusa Mk3 0.2mm layer height "Fast" setting. It worked perfectly.

        Then I opened up two windows of Slic3r and modified my 0.35mm layer height configuration for the Mk3. I took all the speeds and other settings for the Mk2 0.35mm "FAST" and replaced them with the Mk2's settings, EXCEPT the first layer specific settings, which I made identical to the Mk3 0.2mm defualts.

        My problem before seemed to be I was using a 1mm wide first layer, running at what may have been too slow of a speed. Getting the first layer fixed - ie back to a Prusa default - seemed to make the problems go away.

        I'd suggest for your figures to try the default settings, if you are having problems try printing form an STL that you know works with the same print settings.
        The Mk3 seems to be way better at doing detail fast than the mk2, and without all the nasty strings that the Mk2 MMU leaves.

        Next up is to try swapping nozzles on the Mk2 and really cranking the speeds up...

        Comment


        • #6
          Ok - my thoughts were along the lines of slowing the head down to increase the detail, but getting the first layer down faster seems to improve adhesion.

          I'm looking forward to getting some experimentation time on the MK3. Can I book a slot in advance? It seems to be busy (literally) all the time.

          Darren

          Comment


        • #7
          Ok, FINALLY I have found a good resource on print head speed vs nozzle size, extrusion width + height. So much information out there that is pure hand waving an BS.

          http://projects.ttlexceeded.com/3dpr...zle_sizes.html

          For the Mk3 figuring out how fast the print head can move is limited by volumetric flow through the hot end, and goes as follows for PLA:
          Max. Speed = 11.5 mm3/s / (Extrusion Width X Layer Height)

          If you try and go much faster the plastic no melty.

          There is a handy chart in that link with all the different max speeds based on nozzle size & layer heights.

          You can even use autospeed settings, but there is conflicting info on whether or not to do that:
          http://projects.ttlexceeded.com/3dpr...lumetric-speed
          https://shop.prusa3d.com/forum/softw...gs--t4335.html

          Comment


        • #8
          Also I'll note that the Mk3 has been running into layer shifts on a few of the prints.

          At this point I am ready to give up on problem solving that machine due to temperatures in the Makerspace. I doubt the plastic tenting is enough keep the print area at room temp, so I don't want to get completely hung up on what is going into the machine when the environment has major issues.

          Comment


          • #9
            The MK3 is printing really well now with the improved files, Garret H - AND the 0.6mm nozzle has dropped the print time by about 33%, which is huge.

            The enclosure seems to be working with the light in there offsetting any drafts getting through the plastic.

            DarrenAndSherry I purchased the MK3 to relieve the backlog on the MK2 which was due in part to my prototyping mission - I have to catch up on those objects after the delay in getting the printer up and running. I have to pretty much run it constantly for a few weeks to get caught up, justify the purchase and make my deadline! However, let me know what works for you in terms of time needed and when, and lets book you onto the machine. I imagine you will use the 0.2 nozzle then?

            Comment

            Working...
            X