+Tony Blewitt Good point. Some day, I'll try this on the X and Y. So far, I've been happy with the parts I've made on the mill - it holds size and my circular pockets seem to be round - at least I can't detect any problems using calipers. The next time I do a circular pocket, I'll pop a bore gauge in there and see how good it really is.
That's great I have CNCed my G0704 and I am very Happy with it. I only used Ball screws on the X&Y ...the Z axis is the original lead screw ..I don't "see any backlash for the reason I stated earlier
To check backlash, go 10 thou down and back up 10 thou to return to position. Now zero the gauge. Now go 10 thou up and 10 thou back down. The error is the backlash. When you make all the moves in the same direction you don't measure backlash.
Given the large weight of the moving part, you are not likely to get a backlash measurement. Lay it horizontal and you might. Or impose a load by pressing on a load sensor, at a level greater than the head weight.
![HIWWHEE | OHM [Official MV]](http://i.ytimg.com/vi/OJJLtTLqLag/mqdefault.jpg)
you won't see much backlash like that the weight of the head is taking up all the backlash
+Tony Blewitt Good point. Some day, I'll try this on the X and Y. So far, I've been happy with the parts I've made on the mill - it holds size and my circular pockets seem to be round - at least I can't detect any problems using calipers. The next time I do a circular pocket, I'll pop a bore gauge in there and see how good it really is.
That's great I have CNCed my G0704 and I am very Happy with it. I only used Ball screws on the X&Y ...the Z axis is the original lead screw ..I don't "see any backlash for the reason I stated earlier
Yeah, he could rig up a fairly heavy spring as a load
To check backlash, go 10 thou down and back up 10 thou to return to position. Now zero the gauge. Now go 10 thou up and 10 thou back down. The error is the backlash. When you make all the moves in the same direction you don't measure backlash.
worD
trU
Agreed. It shouldn’t take all those extensive moves to figure out backlash.
Given the large weight of the moving part, you are not likely to get a backlash measurement. Lay it horizontal and you might. Or impose a load by pressing on a load sensor, at a level greater than the head weight.
Great ballscrew and stepper
This is not a backlash test. This is a repeatability test basically. Almost.
It says backlash compensation in the description. So obviously this is a test with the backlash compensation enabled.
The test requires both - and + travel to your origin 0
This is not a backlash test, but closer to a repeatability test.. and not even that
While machining, the load vectors varry greatly and under such circumstances one can get an idea of real backlash.
Hey Franco, on that ballscrew... how far does the Z axis move for one complete turn of the ballscrew?
Angel ECU 5mm
that is not a proper thorough testing。 is there clearane in ballscrew in z upward direction? it didn't. show。 during milling there is up thrust force。
can you please post the gcode that you used for these test ?
+George Octavian G0G90G54
X0Y0Z3.
Z.7
G1 Z0 F8.
G4 P1.
Z.01
G4 P1.
Z0
G4 P1.
Z.02
G4 P1.
Z0
G4 P1.
Z.03
G4 P1.
Z0
G4 P1.
Z.04
G4 P1.
Z0
G4 P1.
Z.05
G4 P1.
Z0
G4 P1.
Z.06
G4 P1.
Z0
G4 P1.
Z.07
G4 P1.
Z0
G4 P1.
Z.08
G4 P1.
Z0
G4 P1.
Z.09
G4 P1.
Z0
G4 P1.
Z.1
G4 P1.
Z0
G4 P1.
Z.2
G4 P1.
Z0
G4 P1.
Z.3
G4 P1.
Z0
G4 P1.
G0Z3.
M30
Proove with 0.001 dial
Do not test the Z axis for backlash. Gravity eliminates most of it. Completely useless video