Speed dependent "backlash"

I have been doing some maintenance before using my 40x20 Mighty CNC to make some tools I am in need of (fix a small cnc retrofit and build a small cnc router + some other misc stuff).

I swapped in a spare ball screw on my Y axis today so I can send the other out to get rebuilt and did the gibs on that axis using the amp draw method. (12 amps max, set them at 4.5 amps running 100IPM with no load, tightening anything past 2 amps results in immediate apparent backlash but running them that loose results it bad surface finish and noticeable vibration).

I went to check backlash afterward and noticed that moving .01 at 5IPM gives me .0025 on that axis, moving .01 at 50IPM gives me 0. Over shorter distances down to about .005 this holds true.

The previous screw was at .0045 measured at 5IPM and was giving me quadrant marks with backlash comp set that high.

Servo tuning issue? Stick-slip? it's driving me nuts and am going to have to resort to interpolating a bunch of circles and measuring/adjusting BL until it gets close enough. It cuts much quieter now and surface finish is much improved but I would like to have some idea of what is going on.

The machine is not the greatest thing in the world (it's actually a pain in my butt to be honest!) and had issues before I purchased it judging by the fact that it has 3 different brand ballscrews, one of which looks to have been re-balled 2x in 3 months.. but it's what I have to work with.

Not back lash, but rather following error due to poor tuning. Pay attention to the sign of the error (leading versus lagging) your velocity gain sounds like it is too high.

motion guru said:

Not back lash, but rather following error due to poor tuning. Pay attention to the sign of the error (leading versus lagging) your velocity gain sounds like it is too high.

Click to expand...

Thanks man, I will plug my laptop in to the Y drive and see what I can come up with tomorrow.