RS5000 Studio MAPC completes early

I'm starting up a new program:
RS5000 Studio ver 21
AB 5500 Ethernet Servo

In the program , I created a virtual axis to mimic a conveyor belt that exists in the real world. The virtual axis is running at constant speed (MAJ command).

I have a single servo axis that is effectively a wheel over top of the conveyor. It is setup as a rotary axis with a 90 degree unwind reset value (ie as the wheel goes around 360 degrees, the axis position goes from 0-90 degrees 4 times).

When I detect a part coming onto the conveyor, the servo axis, (sitting at 0 degrees) needs to CAM to the virtual axis. The Cam profile should rotate the axis 90 degrees.

Everything seems to work as planned...detect the part, activate the cam, the servo axis begins to rotate at the proper time, the servo axis follows the cam perfectly for the first 5 cam profile entries, ......and then the axis comes to a stop and the MAPC instruction .PC bit goes true even though it is only part way through the cam profile.

It is extremely repeatable (which I'm thankful for), but I can't seem to make sense of it.

I eliminated all other Motion instructions such that I only had the MAPC instruction and it does the same thing (so I'm not triggering another instruction that stops or overrides the MAPC).

Any ideas on why the MAPC would seem to think it's complete when it hasn't come anywhere close to the endpoint? Ideas on how to troubleshoot it further?

attached: Instruction, cam profile, actual trend

Some discovery info:

I changed it to be a continuous cam. What appears to be happening is that my initial cycle is starting at Cam Profile Point #12. So my wheel thinks it's starting it's rotation at 60 degrees into the cycle.

As a continuous cam, all remaining cycles then run through the entire profile.

If my Cam Lock Value wasn't zero, I'd understand this behavior.

I tried creating new MCCP and MAPC instructions with new tags - same result.

After getting a good all day butt-kickin' from my program, I finally figured out that it did not like the negative values that I had in the cam profile (all derived from trigonomic equations).

After offsetting all of the values to positive numbers, and fixing all the issues that I created in the various attempts to troubleshoot, my chances of seeing my wife and kids again are starting to look better.