Rockwell Automation GMLC Reference Manual - Manual (Page 169)

Publication GMLC-5.2 - November 1999
132 Using the Imaginary Axis
This expression compares the absolute value of the difference between
the actual position of the material encoder and the command position of
the imaginary axis (the true position error in this system) to the
Lock_Tol_0 user variable. Lock_Tol_0 is the position lock tolerance
parameter of the feed axis, set in the Initialization module. By using the
position lock tolerance parameter of the physical feed axis for this
comparison, the feed axis responds to changes in its position lock
tolerance just as it would if it were being moved directly. After a dwell,
the feed is repeated indefinitely.
The correction routine is run as its own task (task 1), independent of the
actual feeding motion. A direct incremental move of the feed axis equal to
the following:
Command_Position_Imaginary - Actual_Position_Material
This expression corrects for any slip of the feed axis. This move is
executed continuously based on a timer. Select Wait for Completion in
this move block to guarantee that one correction move finishes before the
next move starts. The value of the timer determines the frequency of the
correction moves and must be set empirically based on the dynamics of
the application. In some applications, no dwell may be necessary—the
correction moves can be executed back-to-back using only Wait for
Completion. By specifying the velocity of the correction moves as a
percentage of the speed of the imaginary axis, you can make the
corrections to occur faster as the line speeds up.
The correction routine runs all the time—even when the axis is stopped—
and thus the corrections are applied continuously as the axis slips rather
than all at the end of the feed. Be aware, however, that using the
imaginary axis increases the CPU Utilization and running the correction
routine as a separate task reduces the throughput of the main program.
See the CPU Utilization chapter in this manual for more information.
Note that the imaginary axis must be set up identically to the feed axis for
this technique to work properly. In addition, all axes must use the same
user units to allow cross-axis position computations. In
SLIPCOMP.GML, Axis 0 is the feed axis and Axis 1 is the material
encoder. The imaginary axis is the imaginary axis.