Welcome to LinuxCNC.org

I found a quick workaround for now to the spindle issue.
Adding:

s1 m3
g4 p1
s6000
g4 p2

Seems to allow the controller to rearm as the signal "comes back" and since 1 rpm is in the deadzone it's ignored and treated as a valid 0 rpm arming signal.
Then the rpm can be increased and a short delay is smart to let it spool up.

I have to do a trick like that on my BP 308 VMC... a bunch of sxxx with a short g4 between as the caps are shot on the power supply and won't take s6000 from a dead start.

Glad you got it sorted out for now.

John

I think I should increase the diameter of my leadscrews a bit, and I need to support them on both sides.


But it works

Looking good!

John

Thanks.
I was at the local hw store just now and scored some 8mm precision threaded rods that should be better for feeds.
M8x1.25 should be better than M5x0.8 seeing how I have plenty position-accuracy vs stiffness in the design and could use the extra speed.
Also I got perfectly straight rods so no more whipping. Especially when I support em in both ends.

I also got some more MDF so I can make a longer Y table. The plan has always been to use a longer Y table if I need to, and it seems I need 300mm range on Y while my design only delivers 170 now.
Due to the flying X/Z I do have better range there than traditional gantry design though.
I can use entire width of table in X, It also allows tool length sensor to be located outside the main table area to be always available regardless of Y position. (Saving repositioning time when changing tools.)

I also have started work on a ultrasonic touchprobe that will allow contact probing without pressure and should as such yeld very accurate probing results even on soft materials.

What is the hysteresis and repeatably of the ultrasonic sensor?

John