JR North wrote:
Please do tell us how exactly you compensated for the indexing error
in your X/Y spindle position to claim base accuracy to a couple 10's
in your hole positions.....

JR North:

That's an honest question. First, this was done on Fadal 40X20 CNC.
The table positioning accuracy is listed as being .0002, and the
positioning repeatability is listed as being .0001, see below for
specifications:

http://tinyurl.com/6qyog5

Now Fadals have an M46 code that is called "Positive Approach".

================================================== ======================
M46 Positive Approach:

This code causes the machine to move the X and Y axes in a negative,
then positive, direction before the execution of a fixed cycle
(G73-G89). The X and Y axis move .015 in the negative direction at 25
inches per minute, then move .015 in the positive direction at 3 inches
per minute. This returns the machine back to the original location and
then the fixed cycle will execute.
================================================== ======================

What this feature is designed to do, is essentially remove the
influence of any backlash in the ballscrew axis drives. Although,
there is usually very little backlash in a ballscrew system, and any
that there may be is compensated for in the control. The backlash
correction amount is input in the control after doing a backlash
adjustment (which should be done every year or so depending on how
heavily the machine is used). Any backlash on an axis can be easily
checked by positioning your tenth indicator against an object and
moving the axis back and forth with the handwheel, the machine
compensates for backlash almost immediately.

Anyway, I virtually always use positive approach in all my
drilling/boring cycles.

The hole drilling and checking in my little test should fall within
the positioning repeatability range. Could I have been off by a tenth
or so? Perhaps.


--
BottleBob
http://home.earthlink.net/~bottlbob