Roger,
He tried 20 up/down cycles each at 15 IPM, 30 IPM and 60 IPM (60 cycles total). I have attached what the OP prepared showing the results. An increase of .0005" on the down stroke in the 10th cycle at 30 IPM is a bit odd, but it remained constant for the last 10 cycles. I'm not sure I would consider it a significant anomaly, since is consistent for the other 40 cycles, and it's only .0005". I'm guessing that the .0005" change had more to do with the dial indicator than with the machine. I have one that has .0005" resolution, but it's not accurate to .0005".
Do these results meet your suggestion , or do you endorse giving it another go, only doing 50 continuous cycles at a fixed velocity? Do you have a velocity recommendation? As noted, when the problem showed up, he was running Z at only 5 IPM. Maybe he should run at that speed to simulate what happened during actual production?
Gary