Jaycar Delta 3d printer!

Discussion in 'Hobby Engineering' started by rainwulf, Feb 27, 2016.

  1. grandinferno

    grandinferno Member

    Joined:
    Dec 5, 2007
    Messages:
    15
    CQGLHyperion I feel ya. I ran into almost all the problems you did (including ripping the tape off the bed, that I too thought was just protection).

    Those replacement bearings need new carriages built to hold them. My design is somewhere in this thread. If you're talking about the LMU88s.

    Good luck with it man! I've got mine printing at around 120mm/s but I am having problems with jamming. I think it's a combination of the extruding motor being a bit stressed from being jammed so much (I replaced everything else but the motor) and the actual nozzle being messed up from a little too much from over exuberant cleaning. Once I get those little details happening I think it will be fine.

    I've mentioned it earlier in the thread I think, but the Ikea Lack table hack is really good to keep this printer stable (which is the reason I can print so fast).
     
  2. CQGLHyperion

    CQGLHyperion Member

    Joined:
    Feb 3, 2003
    Messages:
    1,433
    Location:
    North Rockhampton
    Cheers for the info. I am looking at printing those carriages you posted earlier in the thred. Is PLA ok or should I go to ABS for those?

    Ikea is a little too far away for me to get to. Might have to grab one next time I am close to a store.

    Another thing I didn't like was how loose the calibration screws in the top of the carriages were. I mean the printer has a lot of vibration / movement when printing! Ended up using thread tape to get a decent amount of tightness to them.

    I have noticed some people use springs to keep them in place. Just need to find where to get them!
     
  3. grandinferno

    grandinferno Member

    Joined:
    Dec 5, 2007
    Messages:
    15
    I just printed mine in PLA. They seem fine.
     
  4. CQGLHyperion

    CQGLHyperion Member

    Joined:
    Feb 3, 2003
    Messages:
    1,433
    Location:
    North Rockhampton
    Had my first go at printing one of your carriages last night.

    First problem is I have a leaky hot end. Looks like I didn't ensure everything was screwed in as far as it needed to be and hot PLA was leaking out the top and bottom of the heating block and dripping onto the print.

    The end print wasn't actually too bad. Problem is the bearing don't fit! They are too wide (15mm outside diameter) which means even if I am able to force them in they won't line up with the slides properly.

    Anyway I have to clear off all this PLA from the hot end which is going to take some effort before making another attempt.
     
  5. grandinferno

    grandinferno Member

    Joined:
    Dec 5, 2007
    Messages:
    15
    I'm going to guess that your PTFE tube wasn't pushed in far enough. It needs to be right in there, otherwise plastic will leak out the top.

    Good luck eh?

    Did you get the LMU88 bearings? If not, which ones did you get?
     
  6. CQGLHyperion

    CQGLHyperion Member

    Joined:
    Feb 3, 2003
    Messages:
    1,433
    Location:
    North Rockhampton
    Wasn't coming out of the top. The nozzle wasn't screwed up against the threaded part on the other side of the heat block, and since there was a gab it leaked out top and bottom through the thread in the heat block. The block was covered in overheated PLA. So when cold it was all stuck together! I had to heat the extruder on manual to get it apart. Suffice to say the actual heating element is covered in burnt PLA which will require me to reheat it again and clean it off somehow?

    Anyway good chance the nozzle is stuffed anyway so I ordered a complete extruder kit.

    Due to those factors above the print came out pretty bad with respect to dimensions and such hence why the bearings wouldn't fit the print.

    http://www.makerstore.com.au/product/linear-bearing-∅8mm-lm8uu/
     
  7. grandinferno

    grandinferno Member

    Joined:
    Dec 5, 2007
    Messages:
    15
    Ah ok. Bad luck dude. :(
     
  8. eviladrian

    eviladrian Member

    Joined:
    Sep 18, 2001
    Messages:
    866
    Location:
    Sydney
    Hey all, I'm probably the only one dumb enough to keep using this printer, but if not then I'd recommend you look at changing out the motor controllers for newer ones.
    I got these TMC2208s online: https://www.aliexpress.com/item/5pc...073.html?spm=a2g0s.9042311.0.0.3da24c4dUfxqQ0

    They reverse the motor direction so you have to edit your firmware, but the difference in noise levels is incredible! They should have put these controllers in from the start.
     
  9. CQGLHyperion

    CQGLHyperion Member

    Joined:
    Feb 3, 2003
    Messages:
    1,433
    Location:
    North Rockhampton
    This reminds me I finally printed out some of the LM8UU carriages on another printer. Time I get around to doing a test fit and test print using them and the replacement hot end I purchased all that time ago!

    Might have to buy some of those drivers once it is up and running again.
     
  10. CQGLHyperion

    CQGLHyperion Member

    Joined:
    Feb 3, 2003
    Messages:
    1,433
    Location:
    North Rockhampton
    Finally printed out my replacement carriages. Designing and printing up so better fan mounts for upgraded / replacement fans for the hot end. Not too far away from putting it all back together.

    Noticed the last version of this printer is on clearance at Jaycar for $240. Good luck finding a store that has one. Sunny coast was the closest (to me) I could find one using the Jaycar website. As expected my local store wouldn't get them to post it up to my local store.
     
  11. eviladrian

    eviladrian Member

    Joined:
    Sep 18, 2001
    Messages:
    866
    Location:
    Sydney
    I've been playing around with layer heights lately, here's a comparison between 0.1mm layers (left) and 0.05mm (right):
    [​IMG]
    It actually works pretty well, although it does seem to get more stringing and blobs at the lower height.

    I tried going to 0.03mm but it didn't really make a difference. 0.03 is on the right in these pics, and you can see the vents on the back are more defined but the surface is generally rougher:
    [​IMG]

    Is anyone else using one of these printers to do minis? Are you getting better results or is this the best I can hope for?
     
  12. CQGLHyperion

    CQGLHyperion Member

    Joined:
    Feb 3, 2003
    Messages:
    1,433
    Location:
    North Rockhampton
    Doing better then me. Can't get the damn thing leveled. Middle sits lower then the edges which isn't a good thing.

    Minis that size will never be good from a FDM printer. You'll need to move to resin.
     
  13. Butcher9_9

    Butcher9_9 Member

    Joined:
    Aug 5, 2006
    Messages:
    2,097
    Location:
    Perth , East Vic Park
    I hear that sometimes lower Layer Heights sometimes make prints worse not better. Something along the lines of Heat from the nozzle reheating the layers below.
     
  14. Franc

    Franc Member

    Joined:
    Jul 10, 2003
    Messages:
    4,743
    Location:
    Melbourne North Sub
    This - some minis will look ok dependant on the model (less angular that you can sand etc) but resin at the moment is the only good reliable option.
     
  15. CQGLHyperion

    CQGLHyperion Member

    Joined:
    Feb 3, 2003
    Messages:
    1,433
    Location:
    North Rockhampton
    OK I need help with

    I have been setting the bed level to almost perfection adjusting Z home accordingly, fine tuning each tower, checking Z center again, and adjusting smooth_red_offset.

    I have been able to make it so each tower @ Z0 = center Z0. Has taken me a few hours to get it pretty damn good. Yes I am a perfectionist.

    OK so now center Z0 is set I move it to there. I then move it so it moves along Z0 to one of the towers. Moving to tower a tower shows no problem, but when I move it back to center, the hot end crashes against the bed!

    Moving from center Z0(after homing) to the rear tower also causes the hot end to crash into the bed, even though calibrated height at that tower Z0 is fine.

    So when I do a circle test print, the first layer is only printed half way across the bed before the nozzle is flat against the zapton tape and no filament can exit it.

    I have done all my calibration with a heated bed. HALP!
     
  16. eviladrian

    eviladrian Member

    Joined:
    Sep 18, 2001
    Messages:
    866
    Location:
    Sydney
    You said something about changing the carriages to suit new bearings, is it possible that the geometry of the arms has changed slightly? I think there's a delta radius firmware setting that can compensate for this bowl/dome effect.

    From a Reddit thread:
    • Test the nozzle height in the center of the bed.
      • If the nozzle height is above the bed surface, increase the horizontal radius

      • If the nozzle height is below the bed surface, decrease the horizontal radius
    If the deviation isn't huge, maybe just set your slicer to make the first layer thicker. When doing those 0.05 and 0.03 prints, I had it start with a raft of three 0.2mm layers first, so even if the bed level isn't perfect the top of the raft is going to be "flat" from the print head's perspective.
     
  17. CQGLHyperion

    CQGLHyperion Member

    Joined:
    Feb 3, 2003
    Messages:
    1,433
    Location:
    North Rockhampton
    Yeah that was my first thought, that I screwed that up. So I started back with the V4 firmware from Jayacar and went from there. Same thing happening. I did have to tinker with the smooth_rod_offset to get the middle to match the towers.

    I built the new carriages on the same dimensions as the originals in the positioning of all the important bits. Mind you it was my first time designing something complex (complex to me) with Fusino 360. Had a look at the design late this afternoon and it really is a mess I made. But they do what they are meant to. Going to refine them a little more so I might end up going back to scratch.

    One thing I noticed was the rear tower end stop seems to be a few mm different to the other two, and after a few more attempts at getting it all square again I think I might have some trouble with the nema on that tower. Makes what sounds like missing step noise when it gets close to the middle / bottom of the tower. Maybe dodgey rails? Possible driver issue? Something I will look at next time.

    Oh and has anyone tried using Marlin 1.1.9 on this machine?

    Also my inductive probe arrived today. Have to design a mount for it and have a chop at auto bed levelling with it. Anyone else had a go with that? Watched a few vids on youtube about someone that has done it already so going to follow their process.

    G29 come at me!
     
  18. CQGLHyperion

    CQGLHyperion Member

    Joined:
    Feb 3, 2003
    Messages:
    1,433
    Location:
    North Rockhampton
    Update: Found two issues.

    First off the grub screw for the GT2 belt pulley on the Z tower NEMA was not tight. This was causing the belt to skip occasionally which varied the height on the Z tower so that repeatable tower height calibrations was impossible.

    Secondly even after fixing that issue I was still getting the same fault where the nozzle would almost crash into the bed moving towards one tower. So I pulled off the bed and checked the spacers. One of the spacers measured 10.2mm where as the rest were all 10.0mm.

    So using my Aldi Cocoon I printed some new spacers I whipped up in F360 which are 3 pairs connected with a span. This way the span also acts as a spacer. Printed them all at once to minimise any height difference, checked they were good and threw them on.

    Calibrated the towers again to match the new spacers and currently running a test print @ 100mm diameter with raft. Problem has gone away!

    So a little more fine tuning to be done but looking a lot better now.
     
  19. CQGLHyperion

    CQGLHyperion Member

    Joined:
    Feb 3, 2003
    Messages:
    1,433
    Location:
    North Rockhampton
    Was using feeler gauges for setting head height. Worked fine when heated bed was off, but when heated bed was on, it causes the connection to the PC via USB to drop. From what I can tell there is a short generated when power is turned on to heat the bed when the feeler gauge bridges the gap.

    Back to using paper then.
     
  20. CQGLHyperion

    CQGLHyperion Member

    Joined:
    Feb 3, 2003
    Messages:
    1,433
    Location:
    North Rockhampton
    Well after a bit more mucking around it looks like I have a decent level to the bed now. A few test prints of a single layer 120mm disc showed there were still some issues to one side.

    So I decided to try a glass plate that I have borrowed off the second hand Geeetech G2S Pro Delta that I got cheap. Adjusted Z height to match, calibrated each tower and checked the middle, and test printed the single layer disc again.

    Almost perfect now. Great repeat-ability on the bottom layer and no adhesion issues using glue stick. Now just to adjust things to reduce/remove the stringing!
     

Share This Page

Advertisement: