Close



Page 2 of 2 FirstFirst 12
Results 11 to 19 of 19
  1. #11
    Engineer
    Join Date
    Dec 2014
    Location
    Canada
    Posts
    498
    Here are the results of the M48 V4 P10

    READ: Mean: 0.664402
    READ: Standard Deviation: 0.002912

  2. #12
    Engineer
    Join Date
    Dec 2014
    Location
    Canada
    Posts
    498
    so whats considered a good result?

  3. #13
    Engineer-in-Training
    Join Date
    Jul 2016
    Location
    Pennsylvania, USA
    Posts
    255
    Hi

    If you are trying to print a layer height of 0.2 mm, a bed that is good to 0.02 is "ok". Anything at 0.002 on a 0.2 mm layer is well beyond what you can see. As you scale layer height, the limit (obviously) changes. It is not at all unusual to look at a "three sigma" (three times the standard deviation) to estimate how good the result will be most of the time. If you have a very un-level bed, and are taking a lot of points, that figures into the estimation as well. On a dead flat aluminum bed, I may only need three points. On a monster I may take a 5 x 5 array. With three points, the flat plate may well be better overall than the crummy bed, even after 25 point correction.

    Bob

  4. #14
    Engineer
    Join Date
    Dec 2014
    Location
    Canada
    Posts
    498
    Are you referring to the Mean or Standard Deviation?

  5. #15
    Super Moderator Roxy's Avatar
    Join Date
    Apr 2014
    Location
    Lone Star State
    Posts
    2,182
    Quote Originally Posted by adamfilip View Post
    Are you referring to the Mean or Standard Deviation?
    .003mm of standard deviation is good. Especially if that was accomplished with 10 probes. I've been able to get numbers down at .001mm but I have my probe leg designed to not move at all. Considering you are moving a lot of stuff to get the switch to trigger, .003mm is perfectly respectable.

  6. #16
    Engineer-in-Training
    Join Date
    Jul 2016
    Location
    Pennsylvania, USA
    Posts
    255
    Quote Originally Posted by adamfilip View Post
    Are you referring to the Mean or Standard Deviation?
    Hi

    The mean is simply whatever the nominal offset to the probe was. The standard deviation is what tells you about the repeatability. Assuming you did a few dozen trials, the SD looks pretty good. If it was over a small number of trials, you need more data.

    Bob

  7. #17
    Engineer
    Join Date
    Dec 2014
    Location
    Canada
    Posts
    498
    Just updated to Marlin UBL v2 redid the probe test

    nice results! from M48 V4 P10

    READ: Mean: -0.997358
    READ: Standard Deviation: 0.000677


  8. #18
    Engineer-in-Training
    Join Date
    Jul 2016
    Location
    Pennsylvania, USA
    Posts
    255
    Hi

    It looks like your data is pretty good.

    You might want to see if this can be moved to a more suitable sub-forum. This is the MakerFarm section and they don't have any CoreXY stuff.

    Bob

  9. #19
    Engineer
    Join Date
    Dec 2014
    Location
    Canada
    Posts
    498
    sure, but this isnt corexy related really.. I am planning on making a version of this hinged gantry for my Makerfarm i3v 12

Page 2 of 2 FirstFirst 12

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •