I have heard of this "sleeping spinner" problem before, and I have even experienced it myself with one particular Opti-PAC. I had an opportunity to have a few Opti-PAC's in front of me once, and one of them displayed just this behavior.
What I found fixed it was to move the trackball to the P1 Rotary header, and move the spinner to the P1 Trackball header. The problem immediately disappeared. In correspondence with Andy, he told me that priority is not given to a particular header and they should all behave exactly the same, so the reason swapping devices corrected it didn't seem to make much sense.
I have my own theory that on an extremely small percentage of Opti-PAC's (I've heard of only 8 instances so far, 1 being myself), when a Happ trackball is connected to the Trackball header it interferes with a spinner connected to the Rotary header, be it because of dual optic boards on the Trackball header or not, I can't prove that... However, moving the dual optic board device (trackball) to the Rotary header and the single optic board device (spinner) to the Trackball header corrects it.
Please let us know if this "fix" works for you, because it may help others in the future.