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

Unresolved Z-axis problems.

Discussion in 'Troubleshooting' started by Oisin, Sep 21, 2016.

  1. Oisin

    Oisin Member

    Joined:
    Apr 14, 2015
    Messages:
    384
    Likes Received:
    23
    Hi guys. I've come here to try to fix this quite a few times before but never got to the bottom of it since it's a weird issue. My printer never seems to be consistently level. I use Jimbots wonderful IR sensor for autoleveling so I can rule out the nozzle touching the bed and flexing it giving me false z readings during autolevel. Everytime I run the autoleveling before a print, I get different z values despite the bed not changing one bit. I have no idea what is causing this and it has been happening since I got the printer a long time ago.

    Modifications to the z-axis:
    Robo leadscrew upgrades (had to get a second set as the first set was warped. This set is too but not as badly).
    Jimbots IR sensor, so that the nozzle never touches the bed and the autolevelling should be far more accurate.

    My z-offset is currently-1.8mm. Because of this weird issue, it takes maybe 30 minutes to start a print due to me needing to guess the correct z-offset that will match up to what the autolevel will read on the next run. For example: earlier it was 2.5mm, then it was -0.5mm and now -1.8mm. When this print is done it may go back up. Who knows.

    Can anyone provide any insight in to what could cause this? If I could run the autoleveling sequence again and again and get the same z values I would be very happy.

    Here are some examples of autoleveling sequence readings that I took back to back just now:

    1st autoleveling sequence:
    RECEIVED: Bed x: 35.00 y: 50.00 z: 0.75
    RECEIVED: Bed x: 120.00 y: 50.00 z: 0.84
    RECEIVED: Bed x: 205.00 y: 50.00 z: 0.71
    RECEIVED: Bed x: 205.00 y: 140.00 z: 0.69
    RECEIVED: Bed x: 120.00 y: 140.00 z: 1.01
    RECEIVED: Bed x: 35.00 y: 140.00 z: 0.63
    RECEIVED: Bed x: 35.00 y: 230.00 z: 0.64
    RECEIVED: Bed x: 120.00 y: 230.00 z: 0.93
    RECEIVED: Bed x: 205.00 y: 230.00 z:0.67

    -1.8mm z offset here. Print was too far from the bed and detached.

    2nd autoleveling sequence:
    RECEIVED: Bed x: 35.00 y: 50.00 z: 0.77
    RECEIVED: Bed x: 120.00 y: 50.00 z: 0.80
    RECEIVED: Bed x: 205.00 y: 50.00 z: 0.66
    RECEIVED: Bed x: 205.00 y: 140.00 z: 0.64
    RECEIVED: Bed x: 120.00 y: 140.00 z: 1.06
    RECEIVED: Bed x: 35.00 y: 140.00 z: 0.72
    RECEIVED: Bed x: 35.00 y: 230.00 z: 0.65
    RECEIVED: Bed x: 120.00 y: 230.00 z: 0.99
    RECEIVED: Bed x: 205.00 y: 230.00 z:0.53

    z values only slightly different here but I tried the exact same z-offset (-1.8) Print was way too close to the bed and knocked the zebra plate off.

    3rd autoleveling sequence:
    RECEIVED: Bed x: 35.00 y: 50.00 z: -0.19
    RECEIVED: Bed x: 120.00 y: 50.00 z: -0.07
    RECEIVED: Bed x: 205.00 y: 50.00 z: -0.23
    RECEIVED: Bed x: 205.00 y: 140.00 z: -0.24
    RECEIVED: Bed x: 120.00 y: 140.00 z: 0.10
    RECEIVED: Bed x: 35.00 y: 140.00 z: -0.17
    RECEIVED: Bed x: 35.00 y: 230.00 z: -0.26
    RECEIVED: Bed x: 120.00 y: 230.00 z: 0.10
    RECEIVED: Bed x: 205.00 y: 230.00 z: -0.69

    Very different z values by almost 1mm in some cases. No change was made to the bed but I changed the z-offset to -1.6. It was still too close to the bed.

    Here is my starting g-code too since there is something to do with the z-offset in there too:
    G28 ; home all axes
    G1 Z5 F5000 ; lift Z by 5mm
    M565 Z1 ; set the offset for auto-leveling mechanism
    G29 ; run auto-level
     
    #1 Oisin, Sep 21, 2016
    Last edited: Sep 21, 2016
  2. Geof

    Geof Volunteer Moderator
    Staff Member

    Joined:
    Nov 9, 2015
    Messages:
    6,757
    Likes Received:
    2,339
    @jim3Dbot may be able to give some aid. I've never messed with IR sensors. I rock the mechanical switches still.
     
    #2 Geof, Sep 21, 2016
  3. Oisin

    Oisin Member

    Joined:
    Apr 14, 2015
    Messages:
    384
    Likes Received:
    23
    I don't think it's anything to do with the sensor. The IR sensor is great! It was happening for the past two years that I've had the printer.
     
    #3 Oisin, Sep 21, 2016
  4. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,914
    Likes Received:
    7,338
    I don't see those sort of changes with the autoleveling I added to the betas. It is identical to the Stock autoleveling. The numbers are pretty dead on after getting it dialed in.
     
    #4 mark tomlinson, Sep 21, 2016
    Geof likes this.
  5. Geof

    Geof Volunteer Moderator
    Staff Member

    Joined:
    Nov 9, 2015
    Messages:
    6,757
    Likes Received:
    2,339
    and the bed is level? you have manually leveled under the magnets for side to side variance that the x is off from the y (after leveling x to Y). that would lead me to believe a warped bed. (uneven all over) which can happen with our style glass.
     
    #5 Geof, Sep 21, 2016
  6. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,914
    Likes Received:
    7,338
    I can see where the IR version should be more accurate (bed is not being pressed and the carriage does not have to move to 'sense' the bed).
     
    #6 mark tomlinson, Sep 21, 2016
  7. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,914
    Likes Received:
    7,338
    But that shouldn't (in his case) result in different reads for different autolevel runs.
     
    #7 mark tomlinson, Sep 21, 2016
  8. Geof

    Geof Volunteer Moderator
    Staff Member

    Joined:
    Nov 9, 2015
    Messages:
    6,757
    Likes Received:
    2,339
    that is very true. curiousity has the best of me on this one. The IR comes down and "senses the bed" so if its uneven could it not sense a high spot in one run, a low spot the next run? Asking because I dont quite understand how they work and never looked into them but imagine them being like the IR sensors I use at work? "light beam" that senses "X" diameter of the position?
     
    #8 Geof, Sep 21, 2016
  9. Oisin

    Oisin Member

    Joined:
    Apr 14, 2015
    Messages:
    384
    Likes Received:
    23
    I have no idea, since I have no method of reliably checking the level. A dial gauge is in the post at the moment.
     
    #9 Oisin, Sep 21, 2016
  10. Rigmarol

    Rigmarol Well-Known Member

    Joined:
    May 14, 2016
    Messages:
    544
    Likes Received:
    280
    I know nothing about using the IR system so I'm nobody, but, if the auto leveling is going to the exact same positions each time and the bed isn't moved and nothing is touching it, then the numbers should always show the same right?

    If I'm thinking right, it points to what controls the nozzle when it starts to print.

    If I'm off base here then I'm safe to ignore.
     
    #10 Rigmarol, Sep 21, 2016
  11. Oisin

    Oisin Member

    Joined:
    Apr 14, 2015
    Messages:
    384
    Likes Received:
    23
    Can you be more specific on the 'getting it dialed in' part? I've replaced so many parts on the printer that it's hard to see what could be causing this.
     
    #11 Oisin, Sep 21, 2016
  12. Oisin

    Oisin Member

    Joined:
    Apr 14, 2015
    Messages:
    384
    Likes Received:
    23
    That's correct. The numbers should show up as exactly the same, but they're always different. It was that way before I installed the IR sensor too. Before, I put it down to the nozzle flexing the bed and giving me weird readings. Now I can rule that out.
     
    #12 Oisin, Sep 21, 2016
  13. Rigmarol

    Rigmarol Well-Known Member

    Joined:
    May 14, 2016
    Messages:
    544
    Likes Received:
    280
    Then if it were me, I'd be suspecting my z steppers or checking for any slop. But I haven't had the issue so I'm just speculating.

    Hope you find an answer.
     
    #13 Rigmarol, Sep 21, 2016
  14. Oisin

    Oisin Member

    Joined:
    Apr 14, 2015
    Messages:
    384
    Likes Received:
    23
    Yeah I just started to suspect that myself. I'm on my 3rd set of z-steppers and rods now actually so I just assumed that they were fine until I saw the x-axis was slightly misaligned earlier.
     
    #14 Oisin, Sep 21, 2016
  15. mark tomlinson

    mark tomlinson ༼ つ ◕_ ◕ ༽つ
    Staff Member

    Joined:
    Feb 21, 2013
    Messages:
    23,914
    Likes Received:
    7,338
    In my case it was just making sure the switches and the brackets that hold them were adjusted and in the correct positions on the threaded rods/carriage. They should not be moving any, the only movement should be the carriage lifting off of the switch.
     
    #15 mark tomlinson, Sep 21, 2016

Share This Page