Jump to content

burtoogle

Expert
  • Posts

    1,529
  • Joined

  • Last visited

  • Days Won

    19

Everything posted by burtoogle

  1. If you enable the "optimize wall printing order" setting then it already does what you are requesting. At the end of each outer wall, it moves into the print a little before any retraction takes place.
  2. Yes, within one object you can position several infill meshes and each can have its own infill style/density, etc.
  3. Hello @espadan3d, that is because your layer view is using compatibility mode. It's doing that because either your computer's hw/sw are not suitable or you have forced it to use compatibility mode in the cura preferences. If it is the sw, maybe updating drivers is required?
  4. Hi @darkdvd, yes, I attach the gcode. I just printed one of these and it came out reasonably well. The ends of the "foot" are thickened because it is doing a retract there but it's not really hideous. KMMB_SingleWallCURA.gcode
  5. Hi Eric, it's like this... As we have discussed, Cura always prints a thin wall using two passes of the nozzle. The first pass will output a line with whatever wall line width has been specified. The second pass will output from 0% to 100% of the normal extrusion amount depending on how close the passes are. The thinner the wall, the closer the lines will be and the less the extrusion amount of the second line will be. In the extreme, where the flow has reduced to 0%, the nozzle will pass down the wall extruding nothing. Now I can understand how that may not be detrimental in some circumstances but, in general, slowly moving a nozzle along a wall while the extruder is stationary (and no retract has been done) could lead to some dribble on the wall. At the very least, it's a waste of time as the wall could be fairly long. So, to avoid that situation, it's better to convert the low percentage extrudes into travel moves. And that's what I have been beavering away at today. As always, things are not as simple as one would like and it has thrown up some other issues which I am chewing over. Your second model has proven to be really useful in that it made me realise that I could be doing some of the wall order optimization differently to make a better job when thin walls are present. Hope this makes sense.
  6. Hi Eric, thanks a lot for that model. It's perfect for my experimentation and provides plenty of challenges!
  7. Thanks for the model, it appears to be sliced OK. Here's my gcode, you can't print it but at least you could load it into a viewer and see what's happening. KXL_5aa4242a25b1b_LandWhoopsterPrototpye.gcode
  8. You can adjust the top/bottom line width so that a particular region will be filled pretty much exactly but, of course, another region may then have a gap. YMMV.
  9. I printed one of those D shapes and it came out very nicely. Just went round and round 25 times. Hard to get my phone to focus but here's a picture...
  10. Hi Eric, that's great, thanks. It works as expected, in the thin wall section, it omits the second wall and as the wall thickens it starts including the extra walls. What I would also like is an STL which have walls varying in thickness on the same layer (i.e. the thickness change is in the horizontal rather than the vertical). Even better, something that goes from thin->thick in both the horizontal and vertical directions.
  11. OK, I have whipped up some experimental code and can now specify that lines that have a flow less than a certain percentage can be replaced with a travel move. Here's a screen shot showing one of your D shapes. It's printed as a single wall with no "phone dial" movement. In the settings you can see a new setting called Minimum Wall Flow which has been set to 30%. Could you please provide another STL file that has a combination of thin walls and thicker walls and, ideally, a wall that tapers from thin to thick. Thanks.
  12. Hi @Modified_Gemini, do you have a sample STL, I could work on, please? Something that has a combination of single walls and thicker sections. Thanks.
  13. As mentioned above, Cura simply cannot do single wall prints. It has to print the walls as pairs. As a workaround, I will put together a PR that suppresses wall lines where the flow is less than some percentage and if it is merged you will be able to play with that in the future.
  14. The way Cura works at the moment, when not using spiralize mode, all walls have to have an even number of lines. It modifies the flow of the second line to try and compensate for the overlap but the implementation has some issues and it doesn't do a perfect job. Until either the overlap compensation is fixed or the wall printing is altered to not try to print a zero-width second wall, it's gonna screw up one way or another.
  15. Have you tried the spiralize mode that Cura provides? It works pretty well as long as the slope of the sides is not too shallow.
  16. Thanks for the files. I can slice that and get all the walls if I use a line width of 0.38 or lower but I am unsure of the print quality as it relies on the wall overlap compensation performing well which, unfortunately, it generally doesn't. If you disable the overlap compensation, it will try to print two lines where there is only space for one and if you enable the overlap compensation the width of the last wall to be printed will be higgledy-piggledy (technical term, aka pretty random). The bottom line is that Cura really doesn't cope very well with walls that are an odd number of lines wide. Hope this helps.
  17. Could you make the gcode and the model available please so I can study them.
  18. It may be related to https://github.com/Ultimaker/Cura/issues/3197 but I am not sure. If you are using the optimize wall print order setting, does disabling that fix the problem? If so then it is probably the same issue.
  19. Hmm, well, I have a particular part that is printed in PETG using a 0.4mm nozzle and the first layer is 0.4mm thick (other layers are 0.2mm thick) and it prints just fine.
  20. Hi @Tafelspitz, I looked at the end of dowel2.gcode and dowel3_1.gcode and they look very similar. Neither contain any gcode to move the nozzle away from the surface at the end of the print. I would expect to see some end gcode that does that and other stuff like turning off the heat, fans, etc. I looked at some other people's UM3 gcode and a typical end sequence would be like this: M204 S3000 M205 X20 Y20 M107 G91 ;Relative movement G0 F15000 X8.0 Z0.5 E-4.5 ;Wiping+material retraction G0 F10000 Z1.5 E4.5 ;Compensation for the retraction G90 ;Disable relative movement M82 ;absolute extrusion mode M104 S0 M104 T1 S0 ;End of Gcode
  21. Hi, thanks for posting the gcode, I can see there that you are getting a retraction at layer change, so just disable the retract at layer change option and the z-scar should go away. FYI, there is already a mod (https://github.com/Ultimaker/CuraEngine/pull/694) to ignore the retract on layer change if spiralization is enabled but it hasn't yet made it into Cura.
  22. Hi, I looked at the dowel2.gcode and cannot see at the end anything that would press down on the part. I also do not see anything to move the nozzle from the part. Could you please post the gcode created by 3.1 so they can be compared. I am interested to see what is different at the very end of the print. I think there could be some stuff missing from the 3.2 gocde?
  23. Hi, to help understand this could you please post some gcode, I don't really need to see it all, just the gcode for the last few layers would be enough. Please include the settings at the end of the gcode file.
  24. Your display is in compatibility mode and the simulation isn't supported in that mode. Either you have forced compatibility mode in the preferences or your computer hw/sw does not support the required version of OpenGL. I hope this helps.
×
×
  • Create New...