X-axis Accuracy Test
5 Bears Home Homebrew CNC bench mill

It was finally time to actually power stuff up, and make something move, via NC! Let's see, 4 months now? More? I don't know, but it was a long time coming. Would the motors be torquey enough? I know, not the high drama of the turboprop test, but still fun!

Oh yes I must caveat, none of this is faked or staged, these are the actual photos and measurements I took. Remember that the accuracy of even these nice M'toyo calipers is +/- 0.001", and the measurement is a transfer measurement using a bore guage.


Setup for the test: first, basic motion is verified via jog commands using Flashcut 2.0. I won't go into details regarding the trouble I had just to get the motor turning... my fault of course, but there is some complexity in setting things up correctly. The parameters for the ballscrew were entered into flashcut as 5.08 turns per inch, which is correct for a 5mm/turn ballscrew.

The table was moved to the location shown, and the bore guage used to set 0. The calipers, after transfer from the telescope bore guage, were also zeroed.

On the Flashcut window, I zeroed out the relative motion DRO. The jog was set to medium speed, continuous. Clicking the jog button, I moved the table a random amount.


The DRO said we moved 1.2525"...
And sure enough! Time for a little happy dance!

With this ground ballscrew and excellent NSK bearing set, I trust the accuracy of the table more than I do the calipers. I have no doubt the accuracy is dead on, within the lead error of the THK ballscrew.

One last peek at the business end of the X-axis saddle plate and ballnut installation.

From here, we move on to some heavier work, the construction of the Z-axis and its mating to the vertical column.