search

ska-sa / katsdpscripts

Observation and reduction scripts for the MeerKAT project
BSD 3-Clause "New" or "Revised" License
3 stars 2 forks source link

Robustness of interferometric pointing reduction with sluggish antennas #462

Closed adriaanph closed 8 months ago

adriaanph commented 11 months ago

(from Tinus Steyn, 9 Nov 2023)

Please note that I have seen in the past (and thus suspect) that another reason for low elevation coverage below 30degEl is the fact that sometimes, there are certain APs that struggle to lock on target when MeerKAT observes at low elevations. So when we do data analysis, the analyse script picks this up and then for the compound scans where one or more APs could not lock, it gives 0 valid solutions for all the APs, not just those APs that could not lock during that compound scan. So then we have to [manually] remove the data from the entire compound scan before we fit a pointing model.

Suggested modifications to the scripts:

  1. in addition to the current default or specified "ant_list", create a fall-back list of "very reliable antennas"
  2. in cases where there's no valid fit using ant_list, repeat the reduction with the fall-back list
spassmoor commented 11 months ago

@tinussteyn could you send me a data set where this happens so that I can test a fix for this bug.

tinussteyn commented 11 months ago

Sure @spassmoor , can use dataset with capture block ID 1699031481, where m028 had issue locking it seems on some scans. Scan 56

tinussteyn commented 10 months ago

Hi @spassmoor , from the latest pointing datasets I've identified the following APs that frequently struggle to lock on low elevation targets: M001, M006, M007, M017, M028 and M046. I've created a Jira ticket explaining the issues for site to investigate: https://skaafrica.atlassian.net/browse/OPS-4468.

adriaanph commented 8 months ago

@spassmoor & @tinussteyn i have created a PR with a solution to this bug. please review https://github.com/ska-sa/katsdpscripts/pull/463.

adriaanph commented 8 months ago

@spassmoor & @tinussteyn note that i believe this has been resolved with https://github.com/ska-sa/katsdpscripts/pull/463, so i'll resolve this now.