[daip] FITLD dies trying to read eMERLIN UVFITS

[Eric Greisen]

Eskil Varenius wrote:
> Dear AIPS support,
> I'm trying to load eMERLIN UVFITS files into AIPS. This worked fine in
> december, and also in January I think. Now I want to re-FITLD the same
> files but it seems that AIPS have changed. When I run FITLD with the
> default I end with the error:
> FITLD3: SOINI: 'RAOBS' COLUMN NOT FOUND
> FITLD3: SOINI: 'DECOBS' COLUMN NOT FOUND
> FITLD3: CQADD: NOT ENOUGH INFORMATION TO CREATE CQ TABLE
> FITLD3: BT2AIP: ERROR   2 GENERATING CQ TABLE
> FITLD3: BT2AIP: ERROR   2 GENERATING CQ TABLE
> FITLD3: ERROR      2 RETURNED FROM BT2AIP
> FITLD3: Purports to die of UNNATURAL causes
> FITLD3: ebur 31DEC12 TST: Cpu=     38.2  Real=    143  IO=      6202
> 
> Reading the HELP file for FITLD says that the DIGICOR behaviour was
> changed 5th of february 2013. Indeed, this is a difference from when I
> successfully loaded these files in january. But, whatever DIGICOR
> option I choose I cannot manage to avoid the error above. It seems it
> has to do with something else. I have tried looking for help on the
> BT2AIP and CQADD functions, but there is nothing on the web or in
> AIPS. I guess this is a low-level error. I have tried on a few
> different files and I get the same behaviour. I have also tried using
> a timerange to load just a few scans, and I still get the same error.
> I have the midnight job enabled and ran it just now to double check if
> it would fix my problem, but it seems not. Unfortunately I'm stuck
> here at the moment and I'm grateful for any ideas on this matter.

I changed the task to treat the CQ failure as real bur not fatal. 
Previously the delay correction was allowed only for arrays labeled 
"VLBA" as was the digital correction.  That has changed.  A MNJ right 
now should fix the task.

Note that there is a correction that was needed for the VLBA correlator 
which is not needed for DiFX correlators.  FITLD triggers off a keyword 
in the DiFX data files (CORRELAT='DIFX') to avoid that part of the 
digital corrections.

I have no knowledge of the e-Merlin system to know whether digital 
corrections should be made and what the decision about the DiFX avoided 
part should be...

Eric Greisen