[dirtdigger257]

Does anyone know when backlash compensation is used by the G code interpreter?
My problem was when adjusting backlash settings in my init program I could see the changes when jogging the respective axis.
But when I ran test cuts in g code it had no affect even setting to a big value it had no effect when run threw the interpreter
I'am cutting a square and a circle
Using cutter comp

Only after restarting my computer
It would change the backlash amount and rate in Gcode
In case someone else is having this problem restarting the PC fixed it for me

My other question is, what is the proper sequence for Setting backlash, homing and should I home to the negative or positive limit switch?

If I home my system set a work offset then home again it's with in a few tenths usually under .0003 of an inch.

If I restart my PC and home it can be off as much as .002

[TomKerekes]

Hi dirtdigger257,

Does anyone know when backlash compensation is used by the G code interpreter?
My problem was when adjusting backlash settings in my init program I could see the changes when jogging the respective axis.
But when I ran test cuts in g code it had no affect even setting to a big value it had no effect when run threw the interpreter
I'am cutting a square and a circle
Using cutter comp

Only after restarting my computer
It would change the backlash amount and rate in Gcode
In case someone else is having this problem restarting the PC fixed it for me

I'm not sure why that would be. Backlash should be applied instantly within KFLOP as soon as the Init program is executed to set the backlash values and the axis is enabled regardless of what moves the axis. Backlash correction is applied when moving in the positive direction and not applied when moving in the negative direction. A movement of 0.5% of the backlash amount must be moved for it to be applied or removed.

My other question is, what is the proper sequence for Setting backlash, homing and should I home to the negative or positive limit switch?

I wouldn't expect it to matter as long as Backlash correction is always set before homing.

If I home my system set a work offset then home again it's with in a few tenths usually under .0003 of an inch.

If I restart my PC and home it can be off as much as .002

I can't think of a reason for this. You might post your homing program and explain how you are measuring this. What if you home twice after a restart?

[dirtdigger257]

My machine is a Novakon NM200 3 axis mill.

I was using a hiemer 3D Touch probe and dial test indicator
Checking x and y from the solid jaw of my vise it has a smooth surface ground finish.
I realized I was getting errors in measurement from not torquing the draw bar down the same torque every time( +- 0.0015 inches).
With that figured out I've retested over the next couple a days and the worst was .0003

I hooked up a glass scale and did some more testing the Z and Y axis look good over the length of travel.
The X axis has a variation off
.0025 to .001
Must be some wear on the X axis ball screw.
I loosened the gibs checked the coupler and bearings they were all fine.
Thanks for the help
I just needed some reassurance that I didn't have some problem with sequencing homing and backlash.