[daip] Odd warning from KRING

[Chris Flatters]

Shami Chatterjee <shami at spacenet.tn.cornell.edu>:

> Working with KRING on two similar datasets, I get this warning that I'm 
> unable to make sense of:
> 
>
> ella  > KRING1: Adjusting solutions to a common reference antenna
> ella  > KRING1: CALREF: CAN NOT RELATE ANTENNA  1 TO  5 IN SUBARRAY  1
> ella  > KRING1: Some solutions were not adjusted to the common
> ella  > KRING1: reference antenna. This may be because one or more
> ella  > KRING1: antennas were not used. Check your data before
> ella  > KRING1: proceeding.
> ella  > KRING1: Appears to have ended successfully
> 
> Both datasets produce this warning for antenna 1 when using refant=5; 
> though the final solutions look very similar when overplotted with those
> from FRING. 
> 
> On my current dataset, refant=9 produces the same warning for antenna 1; 
> while setting refant=1 does not say anything about adjusting to a common
> antenna at all. (Bizzare! Is there a hardwired internal refant=1?)

In this case, this is spurious and results from the way in which KRING handles
the deleted IF. Rather than marking all solutions for the IF bad, KRING sets
the reference antenna to 1 for that IF, sets the delay for antenna 1 to 0.0
and blanks the solutions for antennas 2, 3, and so on.

When KRING comes to re-reference the solutions to antenna 5, it sees that
antenna 1 has a solution referred to itself and tries to find the delay
difference between antenna 1 and antenna 5 for this IF so that it can
calculate the delay relative to the new reference; no such data exist and
the re-referencing is determined to be incomplete.

	Chris Flatters
	cflatter at nrao.edu