Kraken probing

Discussion in 'Multi-Extrusion' started by mhackney, Mar 26, 2014.

  1. mhackney

    mhackney Well-Known Member

    Joined:
    Mar 4, 2014
    Messages:
    53
    Likes Received:
    11
    I've been working on probing to help automate calibration of my delta printer. With 4 nozzles at work, seems like some of the approaches of using contact of a nozzle to trigger capturing the Z position may not be the best approach with Kraken since, in theory, all 4 nozzles would hit at the same time resulting in an effective contact spot of 20mmx20mm. I'm wondering if anyone has thought about this problem and have any solutions in mind or actually tried?

    It seems that the ideal approach would be a probe at 0,0. I don't mind having a Z offset for the probe, this is common for CNC milling probes and is easy to determine and calculate true Z. I also don't mind a manually installed/deployed probe since this is going to be infrequently used and there is no reason to carry the extra mass around.

    Cheers,
    Michael
     
  2. Josh

    Josh Administrator
    Staff Member

    Joined:
    Feb 26, 2014
    Messages:
    183
    Likes Received:
    38
    Hi Michael,

    Its quite a complicated problem! Bed levelling is one thing with a single nozzle, because if the bed is not flat it can be compensated for in Z as you move around the surface of the bed. So you are just fixing a point to a plane..

    ..However with the Kraken, you are not dealing with a single point - you are dealing with 4. If they are configured correctly they will all lie on a flat plane. So in this case you are trying to make 2 planes coincident which is impossible without the ability to rotate the plane that the nozzles lie on (by rotating the entire Kraken I suppose..).

    I am going to stop there before I open up a massive can of worms involving some very scary forward kinematics. :p
     
  3. mhackney

    mhackney Well-Known Member

    Joined:
    Mar 4, 2014
    Messages:
    53
    Likes Received:
    11
    Agreed about the complexity! I think the best approach is to probe with a single point of contact to profile the table. On a delta the trend is to automated calibration of delta radius and the other parameters that lead to fluctuations in Z during X-Y movements. So it is more important to have a small contact area to accurately calibrate the bed.

    So I am thinking a single probe positioned at 0,0. This is very analogous to a machine probe - you probe once then use the tool with tool offsets.

    cheers,
    Michael
     
  4. Josh

    Josh Administrator
    Staff Member

    Joined:
    Feb 26, 2014
    Messages:
    183
    Likes Received:
    38
    That's a good point - I suppose if you know that your nozzles and bed are both level (at least relative to each-other) then the only thing left to solve for is delta-radius, which is easily done.

    Keep us updated on your design for the mount, would be interested to see how you implement it.
     

Share This Page