Z-Axis suddenly out of whack

Discussion in 'Calibration, Help, and Troubleshooting' started by Henry feldman, Apr 7, 2016.

  1. R Design

    R Design Well-Known Member

    Joined:
    Oct 7, 2015
    Messages:
    944
    Likes Received:
    190
    Some time ago I did have a moment of weirdness with elements that are familiar here.

    It came when I'd just installed the Volcano and the IR sensor got moved around.

    I THINK it might have been because, during the belt levelling of the IR sensor, I didn't push the IR sensor hard against the belt, but on the contrary let it ride up a little bit higher (thinking this might reduce potential for accidents).

    The consequence of this was that my Z offset turned out to be slightly positive (rather than negative) but I got confused and didn't REALISE what had happened. I don't know why but there was an hour or two of nozzle above the bed, sometimes by amounts that were far greater than the Z offset confusion.
     
  2. Miasmictruth

    Miasmictruth Well-Known Member

    Joined:
    Sep 4, 2015
    Messages:
    804
    Likes Received:
    118
    If it makes you feel any better for some reason I have to reset my bed level ever other time or so I turn my BigBox on it seems..

    What I have done now, I used a dial indicator and indicated the bed in all 4 corners. I find the first point the set the z offset of the other 8 the same make quick work of the leveling process and works great.
     
  3. Henry feldman

    Henry feldman Well-Known Member

    Joined:
    Sep 5, 2015
    Messages:
    969
    Likes Received:
    391
    Ah, so same as me. Every time I print I have to recommission, but have no idea why... And of course this means my chance of ever getting "dialed in" is 0, since each time you make a tiny error and you print in space or grind the glass...
     
  4. Alex9779

    Alex9779 Moderator
    Staff Member

    Joined:
    Sep 4, 2015
    Messages:
    2,411
    Likes Received:
    735
    That cannot be. Describe your whole workflow after you leveled your bed...
     
  5. Miasmictruth

    Miasmictruth Well-Known Member

    Joined:
    Sep 4, 2015
    Messages:
    804
    Likes Received:
    118
    I know you were directing toward Henry, but for me I may just start a print or home. Generally if I cycle power or come back the next day or sometime just later in the day it will randomly me a 1-2mm above the bed. I don't think much of it just re level but its happened may times now.
     
  6. Alex9779

    Alex9779 Moderator
    Staff Member

    Joined:
    Sep 4, 2015
    Messages:
    2,411
    Likes Received:
    735
    So do I understand you correct that you print after leveling? Or home?
    You have to save right after leveling using M500 or the options in the Control menu... Otherwise your leveling is lost after a reboot...
     
  7. Miasmictruth

    Miasmictruth Well-Known Member

    Joined:
    Sep 4, 2015
    Messages:
    804
    Likes Received:
    118
    Ah, I must have missed the part where you need to do a save procedure.
     
  8. Henry feldman

    Henry feldman Well-Known Member

    Joined:
    Sep 5, 2015
    Messages:
    969
    Likes Received:
    391
    Bring head to center with left nozzle on center dot. Turn off steppers via LCD. Bring bed up to just touch bottom of nozzle. Move head all the way to left limit, check if still just touching tip (adjust as needed with left acme), move to right adjust as needed using right acme. check still level all the way across. Move back to center, redo IR sensor height using cut section of belt. Select Level-Bed on LCD. go through mesh leveling process. control->store memory on LCD. Auto-Home. Print via octoprint. If I got it right, it will print nicely. I then remove said print from bed, and go to print a second job. That job will more likely than not print 1-2mm above the bed. So yes I saved the bed level, and did nothing between the jobs other than upload a new STL to octoprint. According to my stats my jobs are working 43% of the time (i.e. I press cancel 57% of the time, almost exclusively due to bed height)
     
  9. Old_Tafr

    Old_Tafr Well-Known Member

    Joined:
    Mar 7, 2016
    Messages:
    731
    Likes Received:
    93
    Sounds simple but can you say how you used the dial gauge on each corner, for example from which reference point so that all measurements are taken from the same place.

    I found that during the mesh levelling my glass plate sloped slightly towards the back so although compensated it must mean that some Z adjustment is needed to print something that is flat.
     
    #29 Old_Tafr, Apr 13, 2016
    Last edited: Apr 14, 2016
  10. hangsta

    hangsta Member

    Joined:
    Dec 21, 2015
    Messages:
    5
    Likes Received:
    3
    How's your Z motor current setting? If this isn't a software issue then that is the next thing I would check. Really hot motors tend to skip steps. Too much current leads to hot motors, but not enough current also can lead to skipped steps. During printing the Z axis moves so little at a time and problems may not always show up. I found that the Z axis leadscrews really need a lot of lubricant, so make sure they are fully coated for minimal resistance. If that's not it, maybe something is messed up with the IR sensor.
     
  11. Henry feldman

    Henry feldman Well-Known Member

    Joined:
    Sep 5, 2015
    Messages:
    969
    Likes Received:
    391
    They were dialed in perfectly before. I will note that once I have completed this, I can print 30 hour prints with high precision (if I get the z-axis right) [note: the eyes were a flaw in the original model, so had to reprint it after patching the mesh - which again worked perfectly].
    [​IMG]
     
    JohnEsc likes this.
  12. Miasmictruth

    Miasmictruth Well-Known Member

    Joined:
    Sep 4, 2015
    Messages:
    804
    Likes Received:
    118
    Don't really need to hit exact spots, I temporarily attached my indicator to the head of the print head so it couldn't move, then I moved the bed up until I hit roughly the mid range on the indicator.

    Then with the steppers disabled I moved the head as close to each corner as I could. I then slowly turned the standoffs for the high corners down and the low corners up. Just make sure you don't turn them so far they pop off.

    After a few iterations of that I was able to get each corner within a couple thousandths. Oddly enough the middle of the bed seems slightly bowed up but there is nothing to be done about that, or the carriage rods bowed down I'm not sure, either way its not enough to effect the prints much.

    One thing to note as you turn the standoffs make sure you snug the top nut down before you measure, otherwise that corner could be slightly higher then it will be when you tighten it up. If your still confused I can try and do a write up this weekend I will be out of town for the next few days so I won't be able to sooner.
     
    Old_Tafr likes this.
  13. Old_Tafr

    Old_Tafr Well-Known Member

    Joined:
    Mar 7, 2016
    Messages:
    731
    Likes Received:
    93
    I understand now.. thanks !
     
  14. R Design

    R Design Well-Known Member

    Joined:
    Oct 7, 2015
    Messages:
    944
    Likes Received:
    190
    I'd be quite interested to see a photo of how you did that.

    Took a long hard look at my dial gauge, then the print head - and gave up!
     
  15. Miasmictruth

    Miasmictruth Well-Known Member

    Joined:
    Sep 4, 2015
    Messages:
    804
    Likes Received:
    118
    Lol, I cut a couple pieces of thin foam, made a cutout for the back mount of the indicator. I then used a few zipties to secure it as stable as possible by compressing the foam and pulling them tight. Once I was done it was held in place pretty well against the blue fan gaurd.

    If I ever get to redesigning the carraige I will try to design an indicator mount into it somehow.
     
  16. R Design

    R Design Well-Known Member

    Joined:
    Oct 7, 2015
    Messages:
    944
    Likes Received:
    190
    Brilliant...

    Would never have thought of that!

    ;)
     
  17. Old_Tafr

    Old_Tafr Well-Known Member

    Joined:
    Mar 7, 2016
    Messages:
    731
    Likes Received:
    93
    If you ever make a video on how you use the dial indicator then you will get a lot of hits.

    I'm trying to sort out one position (and how) on the X-Carriage that I can mount mine to cover the largest part of the glass plate.

    Most accurate would be to measure at all four corners, but that would mean moving the dial gauge so losing the single reference point.

    The obvious but slightly more complex solution would be to mount two gauges, one left of the X-Carriage and one right of it, then adjust the bed until neither reading changed when moving the X-Carriage to all points of the bed.

    Alex9779 I think made a magnetic mount for the print cooling fans and IR sensor, the dial gauge could be mounted in the same way on the left and right cooling fan positions, saving a re-design just for the gauge. Just print a bracket for the dial gauge to attach to the fan mount.
     
  18. Miasmictruth

    Miasmictruth Well-Known Member

    Joined:
    Sep 4, 2015
    Messages:
    804
    Likes Received:
    118
    I will try to make up something when I get a chance, don't worry about going to exact positions just as close to the bolt as possible with the indicator just genently move the head by and from corner to corner you will probably never get it perfect to the thousandth but within a could for sure.

    Two indicators would kind of defeat the whole operation. Unless you calibrate them both to be exactly the same on exactly the same point.

    The idea is you move to a corner and shoot for a number move to the next and makenit the same number for all 4 corners.

    One thing to remember and you change one the others will change slightly.
     
  19. Old_Tafr

    Old_Tafr Well-Known Member

    Joined:
    Mar 7, 2016
    Messages:
    731
    Likes Received:
    93
    The idea with two dial gauges would be to be able to cover the largest area of the bed without moving either of them. Simply moving the X-Carriage in the X and Y directions to cover the maximum area of the bed . In this way you would in fact overlap in the middle.

    Having two gauges makes no difference in being "the same" the idea being that to get the glass plate totally flat with respect to the X-carriage. This is achieved when neither gauge changes at all, it is not dependent on a specific measurement, simply instead being attached to the same reference point.

    I found that using my "measuring stick" that the mesh calibration was within a few tenths of a mm thus providing an alternative to dial gauges.
     
    #39 Old_Tafr, Apr 15, 2016
    Last edited: Apr 15, 2016

Share This Page