Hello there and thanks for any help that can be offered - This one has me chasing my tail!
Problem in General Terms:
Y axis movement is inaccurate. A move of a known distance, as measured with indicator, is displayed on the controller as something different. The error is much worse when traveling in one direction (Y+) than travel in the other direction (Y-).
Specific Problem Data:
Test-1: With a 2" travel indicator mounted on the table indicating on the spindle, an actual Y axis move of 2.0000" in the negative direction had the controller showing a move of 2.0024". When moving back in the positive direction you would expect to see the same error; however, this was not the case. With the same setup; an actual Y axis positive move of 2.0000" was shown on the controller as 2.0122" This same exercise was repeated 8 times with the errors in -Y holding fairly constant and ranging from .0021" to .0026' and that could simply be my old eyes looking at the indicator and over/under shooting the zero a couple of 1/10'ths. +Y moves were very different, in that, the error ranged from .0117" to .0185".
Test-2: With an indicator mounted in the spindle and indicating on a fixed surface on the table, and both the indicator and controller position zero'd, the axis was moved an arbitrary distance away (say 5"). Then by rocking the axis back and forth and back and back and forth about a half a turn each time on the hand wheel in the .010 per click, the axis was moved to starting point as determined by the indicator. Although the controller position should read .0000 (or really really close), it is now over .100" off. When this same exercise is repeated, the error continues to grow.
What Has Been Checked and Ruled Out:
Encoder - This machine has the older "Red Cap" servo motors (Model: 105 Type: A06B-0315-B032#1001) with 2500 Pulse/Rev encoders (ABS Pulse Coder A860-0320-T112). I honestly thought it was going to be an encoder problem, but after swapping between axes, the issue did not follow the encoder and the unit in question worked fine on the other axis.
Servo Motor Drive Board: - The drive board (A208-1003-0090/02) was also swapped in the same fashion as the encoder with no change noted.
Motor / Ball Screw Coupling - Not slipping
Ball Screw End Play - Way covers pulled back and a ball bearing placed in the center-drill hole of the 10mm pitch ball screw with indicator on that. With aggressive axis movement, indicator did not show axial movement of the ball screw beyond about .0005"
Ball Nut to Table - Not moving
Back Lash: - It is an old machine with a minimal amount of backlash; however, what little lash it has was removed before indicator readings described above were taken.
I know that there are parameters to which changes are made when tuning a ball screw; however, I am at a loss when it comes to an issue like this where the magnitude of the error is 6 or 7 times grater in one direction. Not sure where to go from here.......Fortunately; there are no pressing jobs that have to be run on this machine at this time. So; if there is anyone out there that may have encountered a similar issue with a similar control system (i.e. old Fanuc 10M control) - I would love to hear how it was resolved.
Thanks again and Merry Christmas!!
JV