1. Got a question or need help troubleshooting? Post to the troubleshooting forum or Search the forums!

Solved Ready to smash this pos

Discussion in 'Troubleshooting' started by Joehsmash, Aug 2, 2017.

  1. Joehsmash

    Joehsmash Member

    Joined:
    Feb 9, 2015
    Messages:
    56
    Likes Received:
    17
    so heres the deal, i have now spent the last 3 weeks 5+ hours a night trying to get this first layer thing sorted.

    things i have done to the printer,
    i have an R1 with the threaded rod upgrade kit, as well as a e3d v6 (0.4 installed) also have volcano that i some times swap with. upgraded LED lighting (will be doing RGB mod as soon as the parts get here)

    i have manually tried to level the bed, using paper shims, that didnt work.

    it was suggested to me (by the facebook group) to upgrade to the 1.1.0 to use MESH. Very quickly made the switch. ran through the mesh set up with a 0.2 steel feeler gauge had everything leveled out as best as i can (watched each point to see where the nozzle started to deflect the bed or raise the x carriage) than backed off until i was just feeling a slight resistance. set my z offset and went to print my first layer test print. (https://www.thingiverse.com/thing:2125379). print ended up being rubbish. was patchy, none of the layer lines adhered to the other, it was just a stringy mess. adjusted the bed temp, adjusted the extruder temp, adjusted the nozzle height up and down, and up and down and nothing made a difference. i thought maybe my carriage wasn't level with the bed, so i leveled it as best as i can ( first started with the a 10x10 printed block with my calipers, that helped but i was still needing to make micro adjustments to the z axis as it was printing.)

    than was doing some reading, on the forum, and decided to upgrade to 1.1.4 and use UBL. so i flashed the firmware.

    ran all the required commands, set up a 15 point MESH activated it than did a g26, made the micro adjustments. when the g26 mesh looked well enough to print with. i tried an actual print...

    what
    a
    disaster...

    i have since taken the bed apart and shimmed up the linear bearings ( they were loose in the pillow blocks) taken the x carriage apart, tighened everything up and rebuilt. replaced all of the bearings in the extruder replaced the hobbed bolt. replaced the and re-aligned the hot end. replaced all of the fans, replaced pretty much everything that i can possibly think of.

    im pretty much at a loss for words on what to do with this. I have literally gone through 2kg of filament just printing first layer tests.

    i have since ordered an ir differential sensor it should be here next week. ive ordered a 1/4 inch aluminum plate, as well as 2 10x15 1/4 inch treated glass sheets that i will pick up tomorrow.

    i dont even know what to ask for for help or how to even start with this first layer. everything i try just ends up pissing me off more because nothing works with this thing.

    i did all of this because i wanted to get the perfect first layer. i really want to try and get this printer dialed in so i can start making nice prints again.

    any and all help would be appreciated
     

    Attached Files:

  2. Kilrah

    Kilrah Well-Known Member

    Joined:
    Apr 18, 2017
    Messages:
    498
    Likes Received:
    332
    Don't get the point. Pic 4971 looks perfect. All the others seem not bad enough to cause actual concern either.

    Why? A perfect first layer is useless in most cases anyway. Even with pic 4972 where it's obviously too squished once you have another 2-3 layers on top of that you won't notice a thing anymore.
     
    Geof and mark tomlinson like this.
  3. Joehsmash

    Joehsmash Member

    Joined:
    Feb 9, 2015
    Messages:
    56
    Likes Received:
    17
    Problem is repeatability, I'll get it dialed in like that, but than print it again and it will be completely fubar.

    also in 4972, the layers weren't bonded together, when it was scraped off the bed it was all stranded, pulled apart really easily.

    I have 2 pictures I'll upload tomorrow, same settings just rotated the print, and they're both screwed up but they differ in the ways they're screwed.

    I have an IR differential probe coming now, so hopefully that with the new bed will fix it all.
     
  4. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
  5. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
    If it really is being inconsistent that sounds more like an electronics issue. Perhaps one of the Z stepper drivers (there are two on the R1+) is giving up...
     
  6. Kilrah

    Kilrah Well-Known Member

    Joined:
    Apr 18, 2017
    Messages:
    498
    Likes Received:
    332
    I don't know enough about the R1, but I've just come to the conclusion on my R2 that the bed leveling features can actually cause more trouble than is worth and can be the source of such inconsistencies.
    They might be better for the average Joe than a bed that's all over the place, but as long as you're able to decently level your bed manually you're better off doing so then turning the autolevel and compensations off.
     
    mark tomlinson likes this.
  7. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
    The original Robo (the beta) was totally manual bed leveling and in fact the rails/bed rather sucked even for that so we spent time adapting the hardware to make it easier to do the bed leveling. On the R1/R1+ generally if you can get the X fairly close to level then the autoleveling will pretty much make the rest good enough.

    All that said I still manually level the R1 series as good as possible and one some of the printers I don't even run the autoleveling (G29) any longer.

    With Marlin 1.10 and later MESH can be worth doing since it will account for a warped bed much more effectively, but that is not a version that comes stock.
     
  8. Geof

    Geof Volunteer Moderator
    Staff Member

    Joined:
    Nov 9, 2015
    Messages:
    6,757
    Likes Received:
    2,339
    I'd fault the IR sensor set up. Ditch the IR and go back to manual probe with mesh leveling. Double check your z stepper drivers like @mark tomlinson suggested. If you can get the manual probing to be repeatable (easy enough since it only does it once and always in the same spot) mesh should/will fix your issue. Both my R1+ machines run mesh leveling from @WheresWaldo thread and have no issues with 1st layer.

    Now that said- I dont see in your listing above where you cleared EEPROM after you flashed firmware (M502/M500) if you arnt doing this you are losing the battle before it starts.
     
  9. Joehsmash

    Joehsmash Member

    Joined:
    Feb 9, 2015
    Messages:
    56
    Likes Received:
    17
    i know that i didnt state that in the initial post, how ever i am doing that on every flash. im currently running a 7 point (per axis) with the UBL active and than going to manually adjust the points as closely as i can.

    how ever i just posted in the 1.1.4 forum post what size of feeler is needed to fix up the points that i want to adjust.
     
  10. Joehsmash

    Joehsmash Member

    Joined:
    Feb 9, 2015
    Messages:
    56
    Likes Received:
    17
    i have the R1, so i only have the one driver, all though i have installed the z axis upgrade (direct from robo) its pretty much an R1+ with the exception of the arduino/ramps board ( i have the origional 2 piece ). i had a conversation with buddy from robo today we talked for quite a while on things to do/ fix, 99% of what he suggested i have already tried. with that said i was considering ordering some of the new self regulating steppers as they are a heck of a lot quieter than the ones i have now.
     
  11. Joehsmash

    Joehsmash Member

    Joined:
    Feb 9, 2015
    Messages:
    56
    Likes Received:
    17
    im re-running the G29 P1 right now, one of the issues that i see that i believe the IR sensor will help solve is that when the nozzle probes the bed i can see the bed flex, which means that right away those numbers will be skewed so i guess ill be going back to do all the manual probes.
     
  12. Joehsmash

    Joehsmash Member

    Joined:
    Feb 9, 2015
    Messages:
    56
    Likes Received:
    17
    ran into multiple problems.... but finally got it figured... i submit to you
     

    Attached Files:

  13. Geof

    Geof Volunteer Moderator
    Staff Member

    Joined:
    Nov 9, 2015
    Messages:
    6,757
    Likes Received:
    2,339
    nothing wrong with that, good job. if you would list the problems and your solution so if someone else has the same issues they can read this post and possibly figure it out
     
    Kilrah likes this.
  14. Joehsmash

    Joehsmash Member

    Joined:
    Feb 9, 2015
    Messages:
    56
    Likes Received:
    17
    Once I figure out all of my problems I will make a list of solutions, currently having what I think is an over extrusion issue, will trouble shoot tonight. Also I need to confirm my z axis steps is set correct.

    When I try to print a 20mm test cube it's coming out at 20x20x19 (short on the z axis) also at around 11/12mm there is a section that is really messed up. (See attached photo)
     

    Attached Files:

    Geof likes this.
  15. Joehsmash

    Joehsmash Member

    Joined:
    Feb 9, 2015
    Messages:
    56
    Likes Received:
    17
    That photo was blurry
     

    Attached Files:

  16. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    Z steps are a mechanical and mathematical issue. Also note that most people squish the first layer in search of bed adhesion. So let's say you layer height is 0.2 mm and you squish the first layer by half, so now you print a 20 mm height cube and it only measures 19.9 mm so you think you have an issue. Well you told the printer to print the bottom layer at an actual height of 0.1 mm so you actually got exactly what you asked for, maybe not what you expected.

    Generally if I need that kind of precision I don't squish the first layer of the model using the Z axis. Most slicers will allow you to extrude more plastic for layer one, some call it extrusion multiplier some call it extrusion width, but in each case a slight tweak to that parameter will force more plastic to extrude with the nozzle at the correct height.
     
  17. Joehsmash

    Joehsmash Member

    Joined:
    Feb 9, 2015
    Messages:
    56
    Likes Received:
    17

    I 100% agree with you. How ever if you look at the last photo posted you can clearly see a big issue.

    The problem isn't such a simple one in that instance. My 20mil cube isn't 20x20x19.9, it's 20.2x20.3x19. When it hits around the 11mm distance it looks/seems almost as if there's something holding the zaxis back. All though the steps and the rods are increasing/raising.

    It gives me the impression it just comes to a point where it just gobs out the abs.
     
  18. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,912
    Likes Received:
    7,338
    That is certainly more of a Z axis issue than anything related to steps/mm
    Make sure the rods and bearings on the Z axis are lubricated and moving smoothly and that noting is blocking Z axis travel (like a wire loom getting caught).

    Like @WheresWaldo indicated the Z steps are all mechanical so this is not a misstep in the firmware. It has to be mechanical in nature, something is giving you grief at that spot.

    I am sure with some digging you can get this working.
     
    Kilrah likes this.
  19. Joehsmash

    Joehsmash Member

    Joined:
    Feb 9, 2015
    Messages:
    56
    Likes Received:
    17
    Nothing physical has changed since I went from version 3.0 firmware (robo firmware) I can check. But last tear down I lubed everything.

    Going to adjust my z steps, also calibrate my extruder steps. Than if that doesn't work I'll change the firmware just to verify.

    Before I do yet another tear down.
     
  20. WheresWaldo

    WheresWaldo Volunteer ( ͠° ͟ʖ ͡°)
    Staff Member

    Joined:
    Feb 18, 2015
    Messages:
    5,905
    Likes Received:
    3,593
    I still suggest you NOT adjust Z steps
    Extruder is okay to adjust to your hearts content.

    Like I said it is mathematical there are 200 steps for a single resolution of the stepper. There is a fixed pitch of the threads in the linear stepper lead screws. I believe the R1+PLUS and the upgrade kit uses TR8*4 so the Z steps are 800 per mm of lift. There is no need to adjust that number. It is what it is.
     

Share This Page