Is there a better way to calibrate?

  1. Put calibration card under nozzle
  2. Set Z height to proper friction point where card will just begin to buckle when pushed.
  3. Set jog height to 0.01mm custom increment in Luban
  4. Jog up until loss of friction
  5. Jog down until friction restored
  6. Jog up, counting steps, until loss of friction

In 6) if you jogged up 2 steps until a loss of friction on the card then your backlash is 0.02mm.

Notes:

  1. The above procedure can be optimized, but ensures that the measurement is taken in an up direction immediately following a downward motion, which is important.
  2. The minimum movement the machine will respond to is approximately 0.02mm. Successive 0.01mm movements will trigger motion approximately every other movement (at least in my experience).
  3. If your measured backlash is much more than 0.02mm that will result in software drift due to a firmware bug. Specified manufacturing tolerance is 0.02mm backlash. If you are much above that you have a hardware problem requiring mechanical fixes beyond software capability.

An alternate way to measure, which I have also tried, is to use a caliper to directly measure motion in Z. I have had issues with this, but it would be better to measure backlash over a longer movement, say 25mm, to eliminate possible influence from machine minimum motion restrictions.

6 Likes