Jump to content

burtoogle

Expert
  • Posts

    1,529
  • Joined

  • Last visited

  • Days Won

    19

Posts posted by burtoogle

  1. Thanks for the file. Looking at the gcode you can see that the model is not totally solid. There's an inner ring and the threads are not connected to it. Here's a slice through the gcode that shows what I mean...

     

    Screenshot_2020-04-14_07-38-33.thumb.png.ac68a2550fa03caeb0600f6699e411e7.png

     

    I would have expected to only see walls on the inside and outside of the part with infill between the walls. It looks like a modelling problem to me.

    • Like 1
  2. Hi, when I build Cura on Windows, I run this script first

    
    call "C:\Program Files (x86)\Microsoft Visual Studio 14.0\VC\vcvarsall.bat" amd64 8.1
    
    set INSTALL_DIR=%HOMEDRIVE%%HOMEPATH%\cura\inst
    set INSTALL_DIR_UNIX=%HOMEDRIVE%/Users/%USERNAME%/cura/inst
    
    set PATH=%INSTALL_DIR%\bin;%PATH%
    set PATH=%INSTALL_DIR%\scripts;%PATH%
    set PATH=%INSTALL_DIR%;%PATH%
    
    set PYTHONPATH=%cd%\inst\lib\python3.5\site-packages
    set PATH=%cd%\inst\bin;%PATH%
    
    set CURA_BRANCH_OR_TAG=mb-master
    set CURAENGINE_BRANCH_OR_TAG=mb-master
    set URANIUM_BRANCH_OR_TAG=mb-master
    
    cmake -DCMAKE_BUILD_TYPE=Release -DCMAKE_PREFIX_PATH=%INSTALL_DIR_UNIX% -DCMAKE_INSTALL_PREFIX=%INSTALL_DIR_UNIX% -DBUILD_OS_WIN64=ON -DCPACK_GENERATOR=NSIS -G "NMake Makefiles" -DCURA_VERSION=mb-master ..

    Some of that is specific to my builds (mb-master branches) but the initial call to vcvarsall.bat could be useful to you?

  3. 12 minutes ago, ChaosBlades said:

    So you are saying because it is 3 line widths instead of 4, that is the issue?

     

    Yes.

     

    13 minutes ago, ChaosBlades said:

    Do you know the GitHub issue number, so I can track progress?

     

    AFAIK, there isn't an issue on github specifically about this.

     

    The PR that could form the basis of the solution is https://github.com/Ultimaker/CuraEngine/pull/1210

     

    You can see from the comments there that it still has some way to go before becoming a reality.

     

    • Like 1
  4. Hi, sorry for the slow response, it's been a busy day. So, I think this behaviour has been introduced by a recent revamp of the retraction/z-hop code. Arguably, it's a regression. The only workaround I can suggest is to turn off z-hop. For this particular print, that's of no consequence (I think) because it doesn't appear to do any other z-hops in the print. I recommend you open an issue at the cura github as this should get a proper fix.

     

    • Thanks 1
  5. 1 hour ago, JohnRuthroff said:

    Could Tinkercad be the issue?

     

     

    Perhaps. This often happens if the bottom of the model is not completely flat. Try sinking the model into the buildplate slightly by setting its z value slightly negative.

  6. That's a good point. I don't think I mention that in my README so I'll add a comment.

     

    Personally, I keep all of my configuration files in a git repo and every now and again I make a snapshot so I can recover if anything gets trashed. I have had to rewind a few times due to upstream changes causing breakage but that's the price to pay for living on the edge....

×
×
  • Create New...