[daip] Calib error

[Eric Greisen]

Marina Alexandrovna Kounkel wrote:
> Hello!
> 
> When I try to run calib in one of the fields for the phase calibrator, I get the following error:
> 
> CALIB5: CALREF: CANNOT RELATE ANTENNA  3 TO  1 IN SUBARRAY  1
> CALIB5: CALREF: CANNOT RELATE ANTENNA  4 TO  1 IN SUBARRAY  1
> CALIB5: CALREF: CANNOT RELATE ANTENNA 14 TO  1 IN SUBARRAY  1
> CALIB5: CALREF: CANNOT RELATE ANTENNA 16 TO  1 IN SUBARRAY  1
> CALIB5: Some solutions were not adjusted to the common
> CALIB5: reference antenna. This may be because one or more
> CALIB5: antennas were not used. Check your data before
> CALIB5: proceeding.
> 
> There is noting terribly suspicious looking in these antennas, but I flagged them with tvflg anyway. Upon repeat running of calib, the message is still the same. The images produced for the sources also look terribly terribly out of focus, sometimes with several "ghost" components about them. Is there anything that could be done about it?

What this is saying is that there are times for which antennas 3, 4, 14, 
or 16 were chosen as the reference antenna which most times chose 
antenna 1.  Is it possible that you have done so much flagging that 
there are times containing only a few antennas that are viewed as 
separate times in CALIB?  I have worked hard on CALREF which used to 
report errors that were not real and now it should report real ones 
where it really cannot bring all of the SN solutions onto the same 
reference.  Your imaging results suggest that this is the case.
Note if antennas 3 4 14 and 16 were actually flagged in TVFLG then they 
could not be reference antennas for CALIB to have to process - so TVFLG 
must not have worked.  BUT flagging whole antennas is not needed.  You 
need to identify the times/IFs which fail to have antenna 1 as reference 
- PRTAB of the SN table will show this.  Try to find why this is the 
case and whether some antenna other than these could be specified as the 
reference using the REFANT adverb.

Eric Greisen