[daip] BLCAL, IMAGR, and subarrays

[Eric Greisen]

Rick Perley wrote:
>    Eric:
> 
>    I've found a problem with applying BLCAL, with subarrays.
> 
>    I have a database comprising 4 'subarrays' -- generated using DBCON 
> with four single-source files.  My best map, using all 4 subarrays is 
> pretty good.  Since all four originating files were taken on the same 
> day (this is my 'receiver rotation' experiment), I thought I'd try BLCAL.
>    I get a perfectly good baseline-based solution, using 'SUBARRAY = 0', 
> which according to the helpfile, means 'all'.
>    But when I try to make an image with 'BLVER = 1', only the data from 
> the first subarray is utilized.  Indeed, when I try specify 'SUBARRAY = 
> 2', I get a failure.  Here's the error message:
> 
> IMAGR2: IMACPY: NO DATA SELECTED
> IMAGR2: IMACPY: ERROR COPYING Input UVdata
> IMAGR2: TO UVdata work object
> IMAGR2: Blanking channel    1 image plane
> IMAGR2: OBRGET: KEYWORD CNO NOT FOUND
> IMAGR2: KEYWORD =CNO
> IMAGR2: OBRGET: PROBLEM WITH OBJECT: CLEAN field number 0001
> IMAGR2: IMABLK: ERROR BLANKING CLEAN field number 0001
> IMAGR2: NO DATA FOUND
> IMAGR2: Deleting UV work file:
> IMAGR2: Destroyed UV image file: catno=      1 disk= 7
> IMAGR2: Purports to die of UNNATURAL causes
> 
>    If I then put 'BLVER = -1', IMAGR runs normally.  So I suppose IMAGR 
> is looking for a BL table specific to each subarray?  How do we generate 
> this (and apply it)?
> 
>    Rick

I congratulate you on reading the help file for BLCAL but unfortunately 
it lies.  SUBAR=0 becomes SUBAR=1 in the code.  It would have to be 
restructured to have nearly an outer loop over subarray.  Remember it 
write a correction for baseline 1-2 and that baseline is fundamentally 
different in subarray 1 and subarray 2.  (If it is not, then there 
should not be subarrays.)  You can run BLCAL for each subarray forcing 
the solution into the same BL file with BLVER and then IMAGR should work.

Eric Greisen