J. Kissel

I think it might be time to redefine this FAMIS task. The above trends are only showing the last few days, and 
    (a) any number of things can happen (including several cycles of intentionally aligned vs. misaligned optics), and 
    (b) the +/-10 urad "range limits are not anysort of physical, functionality limits of the oplevs, they're merely "if the optic started at centered when we first started caring, then we'd be alarmed if the optical lever continually zoomed off 'in to the weeds' by 10 urad, or if the alignment was varying over 'days' by 10s of urads.

As the trends above stand, you can't tell either of these worrisome things from these from them.

Admittedly, there are other things to check for that the above trend would show, like
    (a) Has the SUM dropped to some abnormally low value, or
    (b) Does the few day trend show that the optical lever is *not* tracking the intentional alignments and misalignments, or
    (c) Has there been some sort of "sudden" days worth of VEA temperature excursion
but we'd want to see these problems immediately, vs. the "once a week" that this test is made.

I attach a ~1 month ndscope trend, and I see that all of the optical levers (in PIT and YAW) for the QUADs, then all the Triples -- even though they did not *start* as "centered," they haven't drifted more than a few urad since the start of O3B. Thus, I would prefer that we keep the long term alignment reference rather than resetting this zero. Mostly because in the *future* month, we'll see a drop to zero, and we have to take the extra time to figure out if there was a centering that occurred.

Anyways -- just hoping start the conversation, with the goal being *my* usual -- let's think about the regular tests we do, 'cause maybe they're no longer needed and we can stop. OR if we're doing them regularly, let's make sure the results that we post reflect the questions we really want so we can answer them without further inspection. Further, if there are new tools out there that may answer the question better, let's use them (in this case, ndscope templates instead of python script).