Ball Bar Testing

[M. Roberts]

Good morning to all. On a regular basis, we have our vertical CNC mills checked for wear via a Renishaw ball bar. For those who don't know what this is, it is an electronic probe of a known length; spherical on both ends; one in a cup on the table, one in a cup in the spindle. The machine is programmed to move in a circle, in the XY plane, forward and reverse, and an arc in both Yz and XZ. From the data, a graph is charted, showing the actual movement of the machine. That is the jist of it. Now for mt delima; where on the table should these tests be performed?? Common sense (at least to me) would tell you that multiple tests would need to be performed to cover the travels of the machine, and/or in an area of the machine that gets used the most, in the case of a machine that does a lot of repetitive work in a particular area. Am I off base with this? To go further, and using this as an exaggerated example, one runs a BB test on the end of the table, and the graph shows backlash in that area...I do not think that its as simple as adjusting the BL...the "good areas" of the screw would be affected, am I looking at this wrong? Thanks, Mark

 

[Richard King]

Not knowing the name brand, model number, if it is a Box way or linear way machine makes it hard.
You need to start with checking the machines leveling screws and be sure they are tight and then re-level align the machine. Check the gibs to be sure they are set right, what part of the machine has ball screw back lash, if it has back lash. After you do all that, then check with the Ball Bar, the ball bar should be the last test. You don't have to be a rocket scientist to know testing in several locations is better then just one. You may want to post this in the Metrology forum too.

 

[M. Roberts]

Richard,
Thank you for the reply. I should have mentioned that these are HAAS mills...yes, I know, but it is what it is. That being the case, they have linear ways, not box ways. As far as what model, it varies...for sake of argument lets say a VF3. I'm just confirming my thoughts that you need to test the entire travel of the machine, just not a snap-shot....

 

[Richard King]

Richard,
Thank you for the reply. I should have mentioned that these are HAAS mills...yes, I know, but it is what it is. That being the case, they have linear ways, not box ways. As far as what model, it varies...for sake of argument lets say a VF3. I'm just confirming my thoughts that you need to test the entire travel of the machine, just not a snap-shot....

Yes, but double check the leveling of the machine. I have seen machines fail squareness tests when the machine wasn't level. Set 2 levels on the table on X and one on Y and move the table on Y first and it should be with-in .0005" Then check X for the same results, then once you know it is good, then do the Ball Bar.