Positional Tolerance Ballbar

Hi,

My company owns a Haas VF-2 and did a ballbar test (QC2-W ballbar system). I'm trying to interpret the results. Everything looks reasonable except the positional tolerance, which seems high.

Here are some results:
Radius= 5.905512"
Circularity 0.000346"
Positional Tolerance 0.003268"
Scaling Error X 0.098 thou/in

From some google searching, I found this from a brochure:

"In addition the ballbar software will automatically calculate a
positional tolerance for the tested machine. (The positional
tolerance value is an estimate of the planar, bi-directional
positioning accuracy of the machine, within the area enclosed
by the ballbar test, and under unloaded conditions.)"


I guess I have two questions:
1.) How exactly is this true position defined? What is the datum? How much worse will this get if I'm machining at a larger radius?
2.) Is a 0.003" positional tolerance typical for a VF-2? It seems like it's an order of magnitude off from the other errors (circularity for example).

Thanks for your help!

Did you take out the centering error?

Post the results printout and we can help more.

Here's a link to the printout
BALLBAR DATA.pdf - Google Drive

Unless the program does it automatically, I did not. I'm not exactly sure how to do that, let me do some research.

Look at Center offset X and Center offset Y Adjust your g54 by those values and rerun.

I have said this before..... if you have mechanical lash it does not matter what your balbar report is tweaked to.... ballbar servo tuning to some extent is like bondo in dents..... ya still gotta pull as much as you can mechanically.

get in the machine with a dial indicator and some pry bars and make sure she's tight and right. no thrust bearing issues, loose ballnts, ballnut brackets, ballscrew/ motor mounts/ etc. after you do all that laser comp it.... 1 axis at a time bi directional. then if you want to do the ballbar for the iso report fine.... but if everything is right on the bidirectional comp the volumteric will be right, unless you have a gimp motor/ drive.

Oh I see. We had someone come in to run the test. I can't rerun. I think I can approximate it by calculating the offset radius (0.002637") and subtracting that from the positional tolerance (0.003268"). This gives 0.000631" which seems a lot more reasonable.

Yup pretty much, Whoever ran it should have comped that out after the first run. Not sure why they didn't.

thanks both of you for the help/advice

While executing the test you did not insert the actual machine temperature (report states 68 deg, I believe this is not true). Measure, insert and run the test once more.

KBmachine said:

Oh I see. We had someone come in to run the test. I can't rerun. I think I can approximate it by calculating the offset radius (0.002637") and subtracting that from the positional tolerance (0.003268"). This gives 0.000631" which seems a lot more reasonable.

Click to expand...

i know this is old but where did you get the offset radius of 0.002637" from?

Nevermind, I realized where that was after I replied.