[daip] New Ticket - [!QAD-525825]: application of SN delay terms contains unwanted phase wrap

[Huib Intema]

New Ticket: application of SN delay terms contains unwanted phase wrap

This is an issue I came across some time back, when working with Obit on RFI subtraction a la Athreya 2009. Bill's implementation first de-fringes the RFI-infected visibilities using a specially generated SN table, including delay and rate entries. Doing this, I found an unwanted phase offset which could be traced back to the way the the delay rates were applied, a piece of code derived from AIPS:
phase(channel) = phase(reference_channel) + [some_delay_term modulo 2pi] * [channel - reference_channel]
This is incorrect, as it should read:
phase(channel) = [ phase(reference_channel) + some_delay_term * [channel - reference_channel] ] modulo 2pi
I verified this problem does exist in AIPS (SPLIT,SPLAT, etc.) as well as Obit. Problems only arise when reference_channel is a non-integer and some_delay_term > 2pi, so not very often.

It may be that this issue is already resolved, but I just wanted to make sure.


Ticket Details
===================
Ticket ID: QAD-525825
Department: AIPS Data Processing
Priority: Default
Status: Open
Link:  https://help.nrao.edu/staff/index.php?_m=tickets&_a=viewticket&ticketid=451