Jump to content

Gerrardh

Member
  • Posts

    20
  • Joined

  • Last visited

Personal Information

  • 3D printer
    Other 3D printer
  • Country
    AU
  • Industry
    Engineering
    Manufacturing

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

Gerrardh's Achievements

  1. The "native" line width is the diameter of the nozzle - e.g. the width of filament as if it's extruded into free air. If you achieve this width on the build plate, then nozzle is moving at the same speed as the filment is being extruded - no stretch, no squish. You get thicker lines (blue arrow) by extruding faster then you move creating a squish (sideways flow from the nozzle). you get thinner lines (red arrow) by extruding slower then you move creating stretch (elongated filament). In order to achieve the stretch, you require some resistance from the bed (adhesion) and/or adjacent lines of filament, which isn't available if you print inside-out because it hasn't extruded enough material to provide adhesion. I do like the idea of printing the first layer at 150% wider - I will try that. I usually print it thicker, but not wider. Thinking about it more, the two (width and thickness) should increase proportionally.
  2. Hi there, Thanks for the suggestion. I tried a few different settings below, exaggerated to highlight their effects. It's easy enough to make those first few lines larger - potentially large enough to give them sufficient bed adhesion, but you can't make them small enough to dissappear. I realise now why the bed adhesion is a problem with these small lines - The slicer is extruding less than the line width to create a thinner line - effectively stretching the extruded plastic. This requires more adhesion between the extruded plastic and the bed/part, which is not available at this early stage. I've also attached a copy of the project file for reference. LeafSpring_x4.3mf
  3. Hi all, I'm having issue with Cura's printing order. We have options to print 'outside to inside' or 'inside to outside' but neither option is useful. The problem is that 'outside to inside' means that any early defects with bed adhesion (usually the first few mm of a print line) will be visible in the outer skin. However, with 'inside to outside' you end up with the printer making several small extrusions that are intended to fill in very small gaps being printed first - and similarly, if these have any issues with bed adhesion, then they will partially or completely come off the bed. If they come off the bed, they stick to the nozzle only to be deposited somewhere else on the part later. Furthermore, material that sticks to the outer surface of the nozzle eventually bakes hard and causes more problems later (stringing, blobbing etc) - it's important to keep the outer surface of the nozzle clean. Video showing what the opportunities for small defects when printing 'inside to outside': Is there a way in cura to change the print order so that: A) we print away from the inner/outer wall boundary - that is, outer walls and skin print outwards, inner walls and infill (or skin on top/bottom layers) print inwards. B) print 'outside to inside' but with the outer skin last? C) Print first layer 'outside to inside', then subsequent layers 'inside to outside' - this is non-preferred, but it would only leave opportunity for one such defect This problem is super frustrating, I've been fighting it for years hoping that Cura would change something. Thanks in advance.
  4. Found the problem - it was 'Adaptive Layers'.
  5. Hi all, I'm trying to diagnose print issues on a custom built printer, and I noticed that some areas of my print appear to be over-extruded, so I've been trying to find a way to visualise the calculated line cross-section. I found that I can visualise 'Flow' in cura, which is about as close as I can get. The problem is that flow is proportional to speed, so a fast section has a higher flow than areas of lower speed - I really need a visualisation that's independent of speed. Anyway - what I noticed, is that as my layer count increase, so does my flow rate, even though the requested speed and line width are fixed. Why is this? I noticed on my other printer this doesn't happen, so there must be a setting in Cura that I've modified. Any ideas? Flow: Speed: Line Width: Overall Part - notice that the dark red sections are consistent with the over-extruded sections on my prints. Thanks in Advance.
  6. I'm a mechanical engineer - when we do FEA we get a results view similar to Cura's preview. That is, view of the model, heat mapped with a colour scale that indicates different values - be it stress, deflection or in Curas case, travel speed, flow rates etc. One thing that helps immensely in FEA programs is a 'Cursor Probe' - a function that allows us to a) hover over model geometry, and have a display that shows the value on that part of the model and b) place 'flags' at various places of the model so they can compare the values at different locations. Is anyone aware of a function or plug-in that allows users to do this?
  7. Thanks for replying. You might be right about keeping the jerk the same = fewer lines of code, but the problem exists during a print path, so there's no travel involved (refer to attached video). my maximum resolution setting is 0.5mm with maximum deviation 0.025mm. Regards
  8. Hi All, I've built a large CoreXY 3D Printer and I'm debugging problems that I've observed. The machine is running Duet2 Wifi board, slicing in Cura 4.7. I notice that the print head frequently stops and starts on curved surfaces - just momentary pauses, but it's enough to leave behind a small blob - problem is that it does it very frequently so the prints look awful. I've researched a bunch of suggestion and none seem to fit the description all that well. Here are a few observations: The blobs DON'T occur in consistent locations - e.g. a vertical seam, edge or otherwise. Suggesting that the problem is not related to geometry (seam in faces, curve radii, etc). The board is 32bit, so can handle much more complex code without buffer issues. Very simple geometry (straight lines) doesn't have this problem. Looking back over the G-Code for an arbitrary layer, I was looking for changes in the feed rate (assuming that the slicer is introducing momentary speed changes). I noticed groups of M566, G1, M566 commands which momentarily halve the jerk, and speed, before accelerating it again. To be clear- I have no way of testing whether this code relates to movement, but it looks suspicious. The problem didn't occur on my first 2 layers - only higher layers. The first few layers have lower speed, so speed could be a factor some how. Jerk is 12mm/s for print, 24 for travel. Accel is 600 for print and 1200 for travel. The link below shows a video of the problem - the first few seconds (while printing the visible surface) is the most obvious. Any suggestions?
  9. So I decided I would use Change at Z post process to add a lower speed for this particular layer as a work around, after checking the gcode in Cura, I needed to restart. After re-starting, the problem doesn't exist. I now have bridge settings enabled, and the line direction is across the part - not along it. I guess the bug should be updated to "sometimes"? I definitely wasn't working earlier, but now it is. So weird!
  10. FYI, This setting now exists in Cura as "Overhanging wall speed" under the "Experimental" Section.
  11. I've seen others complain about this (in 2019), but I haven't seen bug raised yet - hence the post. Enabling "Bridge settings" defaults to the bridge taking the longest path from point to point, and this is not configurable. I've attached a series of images - I've changed the values for the top and bottom line directions (0 and 90°), I've tried changing the top/bottom line pattern, and changing the orientation of the part - none of these work - as you can see in the images the lines run the full length of the part in each case. When I disable bridge settings the lines run across the part as they should. Please change the default behaviour and/or expose the bridge direction variable for the user to set as desired. Thanks
  12. Request: Very quick suggestion - please make the print settings for multiple extruders have one column per extruder (two extruders, two columns), so the user can quickly compare settings between extruders. Currently state: There is one tab in settings per extruder - two extruders, two tabs. Currently the user needs to switch tabs (and therefore hide the settings for the first extruder) in order to view settings for a second extruder.
  13. That's what I think... while the print head is moving slowly, the fan is blowing on the extruded filament to help it cool... Once it's cool it's strong enough to prevent sagging... There are still limits of course... Each layer needs to be supported somewhat by the layer below, but a single thread of filament WILL support some small cantilevered overhang with minimum sagging.
  14. Hi All, Not sure where to post suggestions, so I'll leave this here - please let me know if it belongs somewhere else. I'd like to suggest a setting that allows the travel speed to decrease (user settable) in areas where the overhang gets above a certain value. I found with my printer, if I left the travel speed at the usual setting (40-50mm/sec) then my overhang tests would only get about 50-55 deg, but if I dropped the speed to about 10mm/s then I could get results much closer to 75-80deg overhang meaning I could print much more complex parts with less support material. Unfortunately printing a whole model (especially a large one) at 10mm/s is out of the question... So it would be super cool if Cura could allow me to reduce the travel speed immediately above areas (for say... 4 or 5 layers) that have overhangs above a certain value in much the same way as supports are added now if overhangs are over a certain value... Hope I'm not the only one who'd find this helpful. Regards
  15. Hi All, I've been struggling with a blockage somewhere in the filament path for a few weeks now - I've tried a whole bunch of things, but I'm not having any luck. Tonight I pulled the bowden tube out, with the intention of cleaning the hot end and cooling block, but I found that the end of the bowden tube had some melted filament pushed back up inside about 5mm or so. I remember when I re-installed it last time after pulling it apart, I pushed the bowden tube back into the cooling block quite a way - until it stopped. I now realise that it doesn't stop until it touches the nozzle which can't be right. So the question is... how far should I install the bowden tube into the cooling block? I'm guessing it should go down to about the heat break, but I'm not really sure if that's enough to prevent melted filament finding its way back up into the bowden tube. Any suggestions?
×
×
  • Create New...