SOLVED Bed level anomaly

Discussion in 'Calibration, Help, and Troubleshooting' started by mike01hu, Sep 1, 2016.

  1. mike01hu

    mike01hu Well-Known Member

    Joined:
    Apr 22, 2014
    Messages:
    957
    Likes Received:
    166
    I am doing a full bed print of shopping trolley coins for a charitable event that I have printed for it before but this is the first time with the BB. All levelling (MBL) was done beforehand then stored and a test piece, then 4 coins printed with very good results. Next I filled the bed with 70 coins and set off and the result of the first layer is shown below where it is clear that the front left corner is low and there is a little front to back variation. I did a live measure of the skirt and instead of being 0.2mm it was 0.39mm(!!!) at the front but it is difficult to get to the back. For now I have micro-stepped Z to compensate for some of it but that is not an answer as it varies. I have suspected this before but this is the largest area that I have used all at once, so has anyone got an idea what is going on?
     

    Attached Files:

  2. Jasons_BigBox

    Jasons_BigBox Well-Known Member

    Joined:
    Mar 25, 2016
    Messages:
    232
    Likes Received:
    69
  3. mike01hu

    mike01hu Well-Known Member

    Joined:
    Apr 22, 2014
    Messages:
    957
    Likes Received:
    166
    It was finally about 0.1mm across the Y axis but having done MBL I expected nothing. The trouble is that 0.1 error with 0.2 layer plays havoc with layer 1 and as the first layer defines the quality of this particular piece it is a bit of a pain.

    Interestingly, the levelling process appears to be working, as the Z drives do adjust when moving the head but after an auto home the nozzle stops at the right height although when I travers it across Y it digs into the bed part way across. I am having thoughts about cable/feed-tube effects on the head when it moves as mine sags towards that part of the bed and could cause minor deflections, but why not when doing MBL, or is it that MBL does not have filament being fed at the same time? I believe that @Alex9779 has a more uniform cable run, so perhaps he an advise me.
     
  4. Jasons_BigBox

    Jasons_BigBox Well-Known Member

    Joined:
    Mar 25, 2016
    Messages:
    232
    Likes Received:
    69
    You're right, all of those things should be dialled out by the mesh bed levelling. It's a bit frustrating hearing people talk about alternative rod/slides on the X and Y axes being better if they deflect less, any static deflections should be removed by the MBL system.

    Dynamic loads during printing can have an effect. Pulling of the cables shouldn't be too bad, if anything it may twist the X-axis.

    I'm really not sure what the loads will be from the extruder. I suppose you could cut a straight length of filament and hold it directly above the head while printing a test and see if it made a difference.
     
    mike01hu likes this.
  5. fpex

    fpex Well-Known Member

    Joined:
    Oct 18, 2015
    Messages:
    621
    Likes Received:
    58
    I find that the problem is and stays the sensor more then MBL. So I have resorted to springs and when I turn on the printer, I redo MBL and adjust the springs at the beginning of the first print.
     
  6. Jasons_BigBox

    Jasons_BigBox Well-Known Member

    Joined:
    Mar 25, 2016
    Messages:
    232
    Likes Received:
    69
    I'm wondering if front to back variation could be the bed tilting slightly under load. If we imagine the screws being a pivot, a small amount of freedom in the rod bearings would allow the bed to seesaw back and forth.

    When levelling the bed we put load on the bed pushing it downwards, if even just slightly. During printing, especially the first layer, there could be little or no load on the bed so will it stay in it's last position? How much does the extrusion process drag the bed behind the nozzle?

    Some food for though. I've seen some very minor y-axis variations which I can see on the first layer but it's not really affecting the prints so I'm leaving it alone.
     
  7. mike01hu

    mike01hu Well-Known Member

    Joined:
    Apr 22, 2014
    Messages:
    957
    Likes Received:
    166
    I found loose screws on the Y bush clamps, so that will have contributed to the problem. These screws are hidden by the frame top and need a short Alan key to tighten without lifting the frame. I was surprised, because I regularly check the clamps but two screws were two full turns loose! There is still a variation front to back but about 0.03mm. I had better results with the second batch with a perfect layer-1.

    You are right about the bed tilt as this is a problem I highlighted some time back due to the short length of the Z-bushes.
     

Share This Page