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

Answered Strange rescaling during print?

Discussion in 'Troubleshooting' started by PropLord, Mar 4, 2017.

Thread Status:
Not open for further replies.
  1. PropLord

    PropLord Member

    Joined:
    Jun 19, 2016
    Messages:
    33
    Likes Received:
    5
    I'm working on the disks on Thor's costume for my cousin. The model is great, good topology, high-poly with a file size of about 2mb. Great. But I've got a problem.

    I cut the model in half so the orientation pics up more detail. The model dimensions are basically 100mm x 100mm. So, in half it becomes 100mm x 50mm.

    The bed levels out correctly (finally!!!) and prints at 200 microns at the default of 210C, and comes out perfect! Except it almost seems compressed. My camera is dead but in the picture I uploaded, this is basically what is happening.

    I recreated the model from scratch in Blender just in case the file was corrupted or something, then re-cut it in Blender (like I do 99% of the time).

    The print itself isn't warped. It isn't even "elephant footing." The bottoms are completely flat. Yet, it comes out an Oval instead of Circular.

    I have printed this and the "new" variation of the model 6 times and I'm getting the same exact problem.
     

    Attached Files:

    #1 PropLord, Mar 4, 2017
    Last edited: Mar 4, 2017
  2. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,915
    Likes Received:
    7,338
    Then one of the axis might be off mechanically. A little belt slack ?
    Something must be affecting the steps/mm on one of them.

    I would first say run a test model like the one I have attached and see if it is completely circular (scale if you whatever size you want to test). It is a one-layer test model.

    If it is circular then you have a model problem, if it is not circular you have a mechanical one.
     

    Attached Files:

    Rigmarol and Geof like this.
  3. PropLord

    PropLord Member

    Joined:
    Jun 19, 2016
    Messages:
    33
    Likes Received:
    5
    I cut them in half, and made them sit flat, like in the pictures. If there was something wrong with the X or Y axis, there should be gapes where not enough plastic was extruded.

    In 1 picture you can see the "gap" required to make a full circle. But none of the detail is missing.
    In another picture it shows them standing with a 3vcr2032 battery laying flat for scale.

    If there was a problem with the Z axis, there would be really nasty bits that didn't print correctly. But it's coming out super smooth, grabs all of the "scroll work."
     

    Attached Files:

  4. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,915
    Likes Received:
    7,338
    Well, let me know if the circle model is actually a circle.
    That will tell the tale. If it is correct then your models are the problem.
     
    PropLord likes this.
  5. PropLord

    PropLord Member

    Joined:
    Jun 19, 2016
    Messages:
    33
    Likes Received:
    5
    Your test model seems to have come out perfect.

    Now, one detail I forgot to mention was I printed the Thor chest Disk models at 10% infill. Could the model be sinking down, and still have perfectly even lines on the outer shell?

    I'm about to print a disk half at 25% infill.
     
  6. PropLord

    PropLord Member

    Joined:
    Jun 19, 2016
    Messages:
    33
    Likes Received:
    5
    The extra density did nothing but waste more filament. When I print it as a full piece it prints just fine. "Basically" a circle. But I have to run it at 100 microns to get the detail semi-properly. As expected, there are too many stair-steps for my liking. Which is why I oriented the model differently. Which is what I do for all of my prints.

    For a test, I made a simple cylinder, cut it in half, stood it up at 90 degrees so it had a flat base to print on, just like the chest Disks. And I'm having the same problem. I don't have a millimeter measuring tool, but I can only assume it lost between 5-10mm in height, while keeping 100% of the detail.

    The dimensions of the model were:
    60mm(X) [wide]
    3mm(Y) [thick]
    30mm(Z) [tall]

    All of the layers are consistent. No X or Y shifting. And there is presumably no shifting in the Z axis as the lines are all properly "glassy" smooth. (I tightened both the X and Y belt as much as possible, no change).

    I printed the cylinder again, but scaled the Z from 30mm to 35 mm, which would make the overall shape 60mm(X) and 70mm(Z). An oval.

    Yet when it printed, it almost makes a perfect cylinder shape (when both halves are joined). So what the heck is going on? I'm afraid to print anything "major" like a 6 part highly detailed helmet or something.
     
  7. PropLord

    PropLord Member

    Joined:
    Jun 19, 2016
    Messages:
    33
    Likes Received:
    5
    I just stumbled upon someone else having the same problem. And they fixed it by doing something with the Firmware. The link that someone else provided on what to do gives a 404 error.
    Page on forum:http://community.robo3d.com/index.php?threads/why-are-my-prints-losing-height-on-the-z-axis.5072/
    "Fix" link, 404'd :http://help.robo3dprinter.com/Wiki/RoBo_3D_Firmware
    OMG... Have all of my prints always been doing this and I just didn't realize it?? How do I fix the Firmware??

    I'm using a ROBO3d R1 Refurbished Unit. I have no idea if it's a Refurbished R1+ or Original. I got it sometime last year I believe.
     
    #7 PropLord, Mar 6, 2017
    Last edited: Mar 6, 2017
  8. Geof

    Geof Volunteer Moderator
    Staff Member

    Joined:
    Nov 9, 2015
    Messages:
    6,757
    Likes Received:
    2,339
    @PropLord if you manually raise your Z (i think I understand this is the axis with issue) does it bind at that layer height or can you take it all the way up and back down again? I doubt...DOUBT...really doubt its the firmware. There just arnt enough of these machines runnign around with that issue (and at one point they all ran pretty similar firmware). Most often its the wiring loom that runs down the right size of the printer or the small wires on the left side (endstops) are snaggin on something causing the z to stutter and not go up
     
  9. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,915
    Likes Received:
    7,338
    Does it have thread rods or leadscrews? That will tell the tale.
    Threaded rods will go into a coupler on the output shaft of the stepper motor (looks much like this: https://www.amazon.com/dp/B00H1LNX5A/ref=sxr_rr_xsim1?pf_rd_m=ATVPDKIKX0DER) and leadscrews are an integral part of the stepper motor so -- no coupler. They are the shaft of the stepper effectively

    Once you know if you have those then no couplers -> R1+ and if you have the couplers it is an R1
     
  10. PropLord

    PropLord Member

    Joined:
    Jun 19, 2016
    Messages:
    33
    Likes Received:
    5
    It's a threaded rod with a coupler believe. I attached some pictures (and a picture of the model number, 0340). There about 21 threads per inch.

    I'm always having to manually adjust the Z-Axis. One side always goes up a tad bit faster. If I print something really tall, the right side is always about .25".

    By manually, I mean I physically grab the couplers and twist them up or down. Do you mean manually in the sense that I tell the Z-axis to go (10mm for instance) up or down in the controls?

    Sometimes it locks and won't allow me to down any further. Which leads me to grabbing the couplers and twisting them until the level buttons are being pressed evenly. Then I have to hit "home" while pressing down on the areas where the Z buttons are so it maintains contact. Because if I don't they get off balance again.

    Which seems to be 1 of the problems the other forum member had. His prints were too short, fixed the firmware, his prints came out correctly.

    So far as I can tell, the wires are not snagging on anything.
     

    Attached Files:

    #10 PropLord, Mar 6, 2017
    Last edited: Mar 6, 2017
  11. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,915
    Likes Received:
    7,338
    That is (at best) an R1 then -- it has the couplers.
    So, if it is an R1 you should not have to worry about the stepper drivers being out of sync, but you may have a stepper motor dropping steps. You can try one of two things (OK, three really):

    1) adjust/calibrate the stepper driver card to boost the current.
    2) test the stepper to see if it is just dropping steps
    3) replace the stepper driver card

    The stepper drivers are little daughter cards that are mounted on the RAMPSD card (which is itself mounted on the arduino).

    Ramps1_4.png

    ramps1-4connectors.png

    ramps-14-diagram.jpg
     
  12. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,915
    Likes Received:
    7,338
    I would suggest as a test that you swap in a stepper driver card from another axis and see if it also loses steps on the one stepper.
    If so it is likely a bad stepper. You could try boosting the current from the driver card (since that is free) :)
     
    Geof likes this.
  13. PropLord

    PropLord Member

    Joined:
    Jun 19, 2016
    Messages:
    33
    Likes Received:
    5
    I must be missing something. I opened up the machine and I don't see anything resembling a card.
    How do I change the stepper driver card?
     
  14. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,915
    Likes Received:
    7,338
    If you removed the bottom cover you should see the stack of electronics under there.
    This is a stepper driver card and they are mounted on the RAMPS board which is mounted on the Ardunio board (like already mentioned).

    Stepper drivers.jpg

    See the pictures above...
     
  15. PropLord

    PropLord Member

    Joined:
    Jun 19, 2016
    Messages:
    33
    Likes Received:
    5
    While I'm looking for a mouse size screwdriver, I printed a cube. When it was done, it was:
    (X) 12/16"
    (Y) 12/16"
    (Z) 10.5~/16"

    Just like the disks, "squashed."

    And by the way, thank you very much for taking the time out to try to help me figure out what's going on.
     
  16. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,915
    Likes Received:
    7,338
    Make sure you have the correct firmware loaded for your printer or the z steps will be incorrect in the firmware...

    Sent from my SAMSUNG-SM-G930A using Tapatalk
     
  17. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,915
    Likes Received:
    7,338
    That could also do this same thing :)

    Sent from my SAMSUNG-SM-G930A using Tapatalk
     
  18. PropLord

    PropLord Member

    Joined:
    Jun 19, 2016
    Messages:
    33
    Likes Received:
    5
    How might I go about doing that? That's what I was asking about earlier.
     
  19. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,915
    Likes Received:
    7,338
    Well, exactly what printer have you got?
    There were (sadly) two flavors of the R! (not R1+, not BETA) and it involved which style of threaded rods you have.
    They can be 5/16 (odd, but there were some like this) or 8mm. The firmware then depends on that.
     
  20. PropLord

    PropLord Member

    Joined:
    Jun 19, 2016
    Messages:
    33
    Likes Received:
    5
    Presumably I have the 8mm rods. There are 21~ threads per inch.
     
Thread Status:
Not open for further replies.

Share This Page