Lots of VMC's allow 2 values for backlash on each axis: one for each direction.  Asymmetrical backlash is not that uncommon.  I have one axis that has 10x the backlash in one direction than the other, so I have no idea what I'd use for a value if trying to backlash compensate without being able to specify by direction.  The reasons for asymmetry are many, but in this case it is the Z-axis which has gravity in one direction and a very heavy head.

Since I understand backlash comp is slated to be moved out of the parallel driver and into the trajectory planner soon, that would be an ideal time to make this change.

Cheers,

BW