Z-Probe granularity



  • I've noticed my z-probe results tend to land only on full step values, is this typical?

    If I run a series of probes, the increments between any two results is also 0.012mm (My step size is 0.0125).

    Just want to be sure that's expected behavior, and if not, what settings I should investigate to change it? I'm using 16i microstepping, but increasing that doesn't change the results I get. This is when doing repeated G30 S-1 on a delta, to check accuracy and repeatability of my piezo probe.

    Thanks!


  • administrators

    The firmware only knows the head position to the nearest microstep. I'm guessing that you have a delta printer with 1.8deg motors and 20 tooth pulleys.

    If you increase microstepping then you may possibly see the resolution improve - it all depends on how friction-free your mechanics is. Doubling the microstepping to 32x reduces the torque per microstep by almost 50%, so you may need 2 microsteps to overcome the friction. Higher motor current may help. The other ways of getting better resolution are to use smaller pulleys (e.g. 16 tooth will improve it to 0.01mm) or 0.9deg motors (which will halve it).



  • Thanks David - sounds like I'm seeing the expected behavior, which is what I was really after.

    I may convert over to 0.9 steppers at some point (and 24v at the same time) but I just wanted to confirm what I was seeing is what I should be seeing. I haven't seen an improvement from increasing the microstepping, it still gives results in the same increments in my tests. That's all fine, as long as it's what it's supposed to do.

    There's only one step worth of variation in my tests currently, at any rate. Here's my last round of testing, which seems pretty reasonable.

    12:28:25 PMStopped at height 0.002 mm
    12:28:22 PMStopped at height 0.002 mm
    12:28:18 PMStopped at height 0.002 mm
    12:28:15 PMStopped at height 0.002 mm
    12:28:11 PMStopped at height 0.002 mm
    12:28:07 PMStopped at height 0.002 mm
    12:28:04 PMStopped at height 0.002 mm
    12:28:00 PMStopped at height 0.002 mm
    12:27:56 PMStopped at height 0.002 mm
    12:27:53 PMStopped at height 0.002 mm
    12:27:49 PMStopped at height 0.002 mm
    12:27:45 PMStopped at height 0.002 mm
    12:27:42 PMStopped at height 0.002 mm
    12:27:38 PMStopped at height 0.002 mm
    12:27:35 PMStopped at height 0.002 mm
    12:27:31 PMStopped at height 0.002 mm
    12:27:28 PMStopped at height 0.002 mm
    12:27:24 PMStopped at height 0.002 mm
    12:27:20 PMStopped at height 0.014 mm
    12:27:16 PMStopped at height 0.014 mm
    12:27:13 PMStopped at height 0.014 mm
    12:27:09 PMStopped at height 0.014 mm
    12:27:06 PMStopped at height 0.014 mm
    12:27:02 PMStopped at height 0.014 mm
    12:26:58 PMStopped at height 0.014 mm
    12:26:55 PMStopped at height 0.002 mm
    12:26:51 PMStopped at height 0.002 mm
    12:26:47 PMStopped at height 0.002 mm
    12:26:44 PMStopped at height 0.002 mm
    12:26:40 PMStopped at height 0.002 mm
    12:26:36 PMStopped at height 0.002 mm
    12:26:33 PMStopped at height 0.002 mm
    12:26:29 PMStopped at height 0.014 mm
    12:26:26 PMStopped at height 0.002 mm
    12:26:22 PMStopped at height 0.002 mm
    12:26:18 PMStopped at height 0.002 mm
    12:26:15 PMStopped at height 0.002 mm
    12:26:11 PMStopped at height 0.002 mm
    12:26:07 PMStopped at height 0.002 mm
    12:26:04 PMStopped at height 0.002 mm
    12:26:00 PMStopped at height 0.002 mm
    12:25:57 PMStopped at height 0.014 mm
    12:25:53 PMStopped at height 0.014 mm
    12:25:49 PMStopped at height 0.014 mm
    12:25:46 PMStopped at height 0.014 mm
    12:25:42 PMStopped at height 0.014 mm
    12:25:38 PMStopped at height 0.014 mm
    12:25:35 PMStopped at height 0.014 mm
    12:25:31 PMStopped at height 0.014 mm
    12:25:28 PMStopped at height 0.014 mm


  • administrators

    I am somewhat surprised that you never see the values from intermediate microsteps. Please confirm that you are using a Duet WiFi or Duet Ethernet, not the older Duet 0.6 or 0.8.5. Also, after setting microstepping to something higher than x16, please run M350 with no parameters to check that the microstepping really has been changed.



  • This is a duet wifi, running 1.18

    I'll test that now and report back.



  • Ok, I set it to 32 steps for x/y/z, homed, and then re-ran my test. I am seeing results smaller than a full step now. Perhaps I'd done something wrong in my past tests, it's been a while since I tried that.

    5:24:18 PMStopped at height 0.012 mm
    5:24:15 PMStopped at height 0.005 mm
    5:24:11 PMStopped at height 0.005 mm
    5:24:07 PMStopped at height 0.005 mm
    5:24:04 PMStopped at height 0.012 mm
    5:24:00 PMStopped at height 0.012 mm
    5:23:56 PMStopped at height 0.012 mm
    5:23:53 PMStopped at height 0.012 mm
    5:23:49 PMStopped at height 0.005 mm
    5:23:45 PMStopped at height 0.005 mm
    5:23:42 PMStopped at height 0.005 mm
    5:23:38 PMStopped at height 0.012 mm
    5:23:35 PMStopped at height 0.012 mm
    5:23:31 PMStopped at height 0.012 mm
    5:23:27 PMStopped at height 0.012 mm
    5:23:24 PMStopped at height 0.012 mm
    5:23:20 PMStopped at height 0.012 mm
    5:23:17 PMStopped at height 0.012 mm
    5:23:13 PMStopped at height 0.012 mm
    5:23:09 PMStopped at height 0.005 mm
    5:23:06 PMStopped at height 0.012 mm
    5:23:02 PMStopped at height 0.012 mm
    5:22:59 PMStopped at height 0.012 mm
    5:22:55 PMStopped at height 0.012 mm
    5:22:51 PMStopped at height 0.012 mm
    5:22:48 PMStopped at height 0.018 mm
    5:22:44 PMStopped at height 0.018 mm
    5:22:40 PMStopped at height 0.018 mm
    5:22:37 PMStopped at height 0.012 mm
    5:22:33 PMStopped at height 0.012 mm
    5:22:29 PMStopped at height 0.012 mm
    5:22:26 PMStopped at height 0.012 mm


Locked
 

Looks like your connection to Duet3D was lost, please wait while we try to reconnect.