Jump to content

foehnsturm

Ambassador
 View Profile  See their activity

  • Posts

    1,821

  • Joined

  • Last visited

  • Days Won

    63

 Content Type 

Everything posted by foehnsturm

  1. foehnsturm
    Both materials are PLA. A PLA / FlexiFil print is one of the next things to try.
  2. foehnsturm
    gcode editing is two simple search & replace (exclude the first few lines with the first T0 command): T0 -> M104 T1 S206 // lower temp of unused extruder M104 T0 S214 // rise temp of active extruder G0 F19000 X102 Y5 // extruder change movement pattern G0 F7000 X143 Y5 G0 F19000 X143 Y165 G0 F7000 X102 Y165 T0 vice versa for T1 is to prevent from substituting any T0 or T1 within another gcode command.
  3. foehnsturm
    https://www.youmagine.com/designs/ultimaker-robot-remix-pack "Created for the Ultimaker Dual extrusion launch." ... which didn't happen. So I launched it by myself :wink: http://umforum.ultimaker.com/index.php?/topic/9657-a-different-multi-extrusion-approach-um-tool-printhead-changer/?p=100809
  4. foehnsturm
    https://www.youmagine.com/designs/ultimaker-robot-remix-pack Looks like no one actually printed it yet :wink: 5cm tall, 3 hrs printing time, 400 extruder changes, no wipe tower, no oozing shield etc. y offset for second extruder still off by some 0.1 mm. First dual extrusion print ever without a failing PSU (which embarrassingly enough was my fault :oops: )
  5. foehnsturm
    It looks like 19V supplies with more power are hard to find. Any recommendations? Should I replace the on board 12V regulator by a step down type to run off a 24V supply? EDIT: I'll proceed as described here http://umforum.ultimaker.com/index.php?/topic/7462-power-supply-for-ultimaker-1/?p=70903
  6. foehnsturm
    I'll stop here. Everything works neatly - until the power supply shuts down ... That's aiming at a moving target. I'll have to get a stronger one.
  7. foehnsturm
    One of the first posts I ever made here talked about FDM as a possible evolutionary side branch. There are already or at least will be methods which are more precise or faster. However, I'd call them all single material processes. I don't talk about color here but about materials with really different properties. As competition for FDM will rise its unique features will be of more and more importance. And combining the wide and steadily growing range of applicable materials is a key feature. This is what led me to the tool change idea: Combining most different materials with most different properties in one piece will require most different tool heads. (filament extruders, paste extruders, lasers ...) However, I know printers of this type almost certainly won't be the entry level mass market machines. But why not use a detachable cartridge system even for single extruder setups?
  8. foehnsturm
    Tool changer and cartridges are working reliably and precisely. But since yesterday I encountered: A ) Sudden power shutdowns. Looks like it happened when both heaters left the PID_FUNCTIONAL_RANGE at the same time and entered max current mode. I compiled a new Marlin with reduced BANG_MAX and increased PID_FUNCTIONAL_RANGE. B ) Extrusion problems -> slightly displaced stepper -> worm gear was loose. Fixed it. C ) Weird z calibration issue -> a nozzle becoming loose. Fixed it. D ) X offset between two nozzles where there shouldn't be one -> Hotend PEEK slightly off center. Adjusted Cura settings. E ) Spare weird x displacements during printing. I've seen that before. Looks like a wire crosstalking issue. No real solution yet. F ) Bad temperature control on one hotend. Discovered that the older one had a longer and stronger heater cartridge. Autotuned both hotends. G ) Numerous Z re-calibrations ... I already knew, double extrusion means double trouble ...
  9. foehnsturm
    With dual extrusion Cura produces gcode like: ... T0 ... path for extruder 1 ... move Z to layer n+1 ... path for extruder 1 ... T1 ... path for extruder 2 ... move Z to layer n+2 ... path for extruder 2 ... T0 ... This saves 50% extruder changes. But is it possible to get the dumb, straightforward sequence with Cura as well? T0 move Z to layer n+1 ... path for extruder 1 ... T1 ... path for extruder 2 ... T0 move Z to layer n+2 ... path for extruder 1 ... T1 ... path for extruder 2 ... I'd like to try if dual extrusion benefits from printing with T0 at the original Z height and T1 shifted by layer height / 2. Like: some tweaking in the first layers.... T0 move Z to layer n+1 ... path for extruder 1 ... T1 move Z to (layer n+1) + (layer height / 2) ... path for extruder 2 ... T0 move Z to layer n+2 ... path for extruder 1 ... T1 move Z to (layer n+2) + (layer height / 2) ... path for extruder 2 ... Can't do that with the standard Cura gcode sequence. If layers are 0.1 mm or finer you won't notice the height difference. But in theory the risk of touching already printed stuff with the second nozzle would be considerably reduced.
  10. foehnsturm
    A wild guess: 5 mm/s is close to the speed limit of the stepper with a 1:40 gear. Perhaps with 0.6A it is not strong enough to perform the movement reliably in both directions. The late extrusion also points me in this direction. Did you try 3 or 4mm/s?
  11. foehnsturm
    The steppers I'm using can take 0,8A so I set the pololu Vref to 0.32 (more or less). They can get warm/hot up to 60-65 °C. Not a big problem for the stepper, but as my printhead is made of PLA this already caused some deformation. ABS or XT should withstand it. Right now, I place a small fan next to the stepper when I'm doing long prints like the Voronoi Yoda.
  12. foehnsturm
    Videos: by the end of the week (at least hoping that). I just used the move axis command on the Ulticontroller. It's a very simple movement pattern, changing from first to second extruder: X102 Y172 ; move to back right cartridge holder X143 Y172 ; slide the cartridge in X143 Y7 ; perpendicular move away, to the front right holder X102 Y7 ; slide out (grab) the cartridge Then I used a simple text editor to replace the dual extrusion gcode generated by Cura: Search for: T1 Replace with: G0 F15000 X102 Y172 G0 F7000 X143 Y172 G0 F15000 X143 Y7 G0 F7000 X102 Y7 T1 The same, with a varied movement pattern, for T0 I didn't replace the T0, T1 commands at the beginning (extruder priming).
  13. foehnsturm
    Yes but it's connected to the heated bed power supply (I forgot to mention).
  14. foehnsturm
    The entire setup is working as it's supposed to :smile: But the second, refurbished Merlin hotend shows heavy underextrusion :sad: Had to order a new one ... BTW: There was a sudden power shutdown one time. I thought the UMO (19V, 6.32A) power supply should be able to handle dual extrusion. (Merlin heater cartridge is 30W, heated bed with separate power supply) Any ideas?
  15. foehnsturm
    With the heads parked in adjacent corners I would say 2 to 3 sec.
  16. foehnsturm
    I think I've got all the parts needed for dual extrusion now. The only thing missing is time ... I'll start with two mirrored NEMA8 direct extruder cartridges. Two steppers for a possible FlexDrive are lying around here as well.
  17. foehnsturm
    Thanks, it should read "Single pass is not naturally better". What I tried to say is that if the part is big enough a double pass is at least on par with a single pass. Perhaps it even has some advantages.
  18. foehnsturm
    Single pass is not better as a matter of principle. But if the nozzle comes back to the same location before the extruded filament cooled down sufficiently, you are in trouble. 0.8 x 0.3 mm2 is a huge cross section, it will take some time to cool down. Single pass with infill (and a minimum layer time setting) will always avoid an untimely return to the same location. With a double pass there is absolutely no way to control that besides print speed: time to return = perimeter length / print speed.
  19. foehnsturm
    I've been using the crossflow fan since I started this topic. I think it's working very well for about 98% of possible geometries. Almost evrything I posted in the gallery was printed with this setup. However, I only print PLA, ABS might be a different story. What I really like about it, is that you get rid of all the bulk on the printhead. It even provides enough airflow to use actively cooled hotends without an attached fan.
  20. foehnsturm
    50% more holding force should be ok because the split/unlock move doesn't require much force.
  21. foehnsturm
    Hi Arjan, Congrats, a first attempt that already works. I don't see that happening often ... I haven't got an UM2 so please forgive me if I'm asking silly questions ... If you replace the long screws with shorter ones would it be possible to mount the metal assembly directly to the (maybe a little thicker) bottom plate of the plastic part? If so and if the bowden fitting could be moved downwards, the thick plate on top wouldn't be needed any longer?
  22. foehnsturm
    Almost exactly my thoughts... The "T" with mirrored cartridges is my first choice for a dual setup. If standard electronics would allow for more a quad setup with "X" shaped gripper and identical cartridges and holders in each corner but rotated by 90/180/270° would be interesting. And yes, you can put the bulk above the gripper. The only thing to bear in mind is that there is not too much inert mass or a long lever where a force can attack to break the holding force of the magnets. But the 80 gr of the NEMA8 were no problem at all.
  23. foehnsturm
    Uploaded sample files for a 30x35mm mount (30x30mm cartridge, 35mm high). Cut out the magnets of you choice (2 x approx 1.5 kg holding force is fine) and incorporate the mounting plates into your design.
  24. foehnsturm
    sorry, no .ipt. IGES, 3DS? Just read somewhere that autodesk inventor may be able to open .3dm (openNURBS) directly?
  25. foehnsturm
    I use Moi3D. Yes, a simple, single-sided gripper would be easier but I can only think of designs here which occupy a considerable part of the print area.
×
×
  • Create New...