tile 8552 at the top of the NGC is raising hexapod limit errors (2024-12-06 and 07)

[sybenzvi]

DESI tried and failed to observe tile 8552 two nights in a row (2024-12-06 and 2024-12-07), with the failure due to a hexapod limit error. The failure wasted about 7 minutes of slewing time as the telescope moved from the bottom of the SGC to the top of the NGC (see attached plot from @araichoor).

The mitigation tonight is "update MTL and hope," and hopefully the changing path of the moon will not force DESI so far to the north for a third time. If NTS picks this tile again, we'll have to get more aggressive. There is no way for 8552 to be manually aborted by the LO so we may need to remove or deprioritize it (and potentially all neighboring tiles). See this discussion thread on #survey-ops.

[sybenzvi]

We've had continued problems hitting the hexapod rotation limit this week. Penalizing long slews between the SGC and NGC failed to prevent the telescope from zig-zagging between high-dec tiles and low-dec tiles in the SGC, so instead we reduced the PRIORITY_BOOST of all tiles with Dec>+70° on 2025-01-03. The change was made to tiles-main.ecsv as follows:

from astropy.table import Table

tm = Table.read('tiles-main.ecsv')
highdec = tm['DEC'] > 70

tm['PRIORITY_BOOSTFAC'].format = '%7.6f'
tm['PRIORITY_BOOSTFAC'][highdec] = tm['PRIORITY_BOOSTFAC'][highdec] * 1e-4
tm.write('tiles-main.ecsv', overwrite=True)

This is a reminder to revert the lowered priority once we've addressed the hexapod rotation limit:

# Restore priority at high dec
tm['PRIORITY_BOOSTFAC'][highdec] = tm['PRIORITY_BOOSTFAC'][highdec] * 1e4
tm['PRIORITY_BOOSTFAC'].format = '%7.3f'
tm.write('tiles-main.ecsv', overwrite=True)