EHT 2023 VLBI DiFX correlation setups
Most tracks were observed at 230 GHz, the single track e23d15 at 345 GHz.
Channel definitions are now sorted in numerically increasing order in the DiFX v2d file. This addresses feedback from EHT 2021 by the L1/CE group, 2021 b1 b2 had decreasing freq order (and b3 b4 increasing order), and this lead to problems in postprocessing.
The VLBI Monitor database contains incorrect (truncated) reference pad name entries for NOEMA. M. Bremer provided the pad info separately by email; tracks g17 e19 a22 pad N007, c16 c18 pad E003).
Coordinates for Sgr A* by V. Fish using position and proper motion from 2022ApJ...940...15X for J2000 epoch 2023.29 are 17 45 40.032073 -29 00 28.26098.
Tracks e23g17 and e23a22 have a few scans that were scheduled for ALMA only with no other stations. Although the scans were recorded on ALMA Mark6, they have no baselines, leading to issues under CASA. Thus the problematic ALMA-only scans are commented out in the DiFX correlation setup.
The SMT 345G receiver is not sideband reparating. Receiver LSB folds onto USB. According to Slack #smt_backend on 12 April 2023, b1 b2 Mark6 recorders had no IF signal. A. Lowitz confirmed that the 345G "USB" IF was routed to b3 b4 Mark6 recorders, and the setup was identical to EHT 2021 and the standard EHT tuning was used.
Use v2d deltaClockAccel for LMT to improve on the non-linear residual fringe rates due to their drifting non-disciplined 10 MHz standard.
Derive better SPT coordinates after correlation, if residuals with a priori SPT coordinates turn out to be too large.
ALMA LO offset 345G likely zero? Or like 2021 after all?
Reason for no SMT 345G fringes?