[daip] IMAGR

[Eric Greisen]

Nectaria Gizani writes:
 > 	Hi, 
 > 
 > I was running Imagr on a very faint source, EVN data. I have boxed the
 > real emission and carried on cleaning. When the new residual map came up,
 > with the source "gone", I have put boxes on what I thought was also real
 > emission and I turned off the tv. When imagr finished after 5000
 > iterations, and I saw the map, it looked fine to me, with a low
 > noise. However I had a look at the message window. It was then that I saw
 > that somewhere in the 300 iteration, imagr reported "something is going
 > wrong, abandon clean". However it carried on for the rest of the
 > iterations ending succesfully, instead of dying!! I include the message in
 > what follows. Is there something to worry about?

   Perhaps there is something to worry about.  Before you turned off
the TV you changed the "BOX" to include all of the image including
edge pixels that are always affected by imaging defects.  I include
some of the relevant messages:
 > IMAGR2 10:44:17 Field   1 min = -906.8 MicroJy,max =  998.0 MicroJy
 > IMAGR2 10:44:17 CLNTEL: TELL operation code CHAN received
 > IMAGR2 10:44:17 Loading field   1 to TV from -1.289E-03 to  3.015E-03
 > IMAGR2 11:08:10 TVDBOX: BOX( 1) =   248.00  249.00  253.00  254.00
 > IMAGR2 11:08:10 TVDBOX: BOX( 2) =   259.00  249.00  265.00  252.00
 > IMAGR2 11:08:10 TVDBOX: BOX( 3) =   264.00  251.00  268.00  256.00
 > IMAGR2 11:08:10 TVDBOX: BOX( 4) =   246.00  271.00  249.00  274.00
 > IMAGR2 11:08:12 Clean continuing
 > IMAGR2 11:08:12 BGC Clean: using 1019 cell beam + residuals >    73.21
 > MicroJy
 > IMAGR2 11:08:12      110 Residual map points loaded
 > IMAGR2 11:08:12 Reached minimum allowed flux   =  320.505 MicroJy iter=
 > 104
 > IMAGR2 11:08:12 Total Cleaned flux density     =    7.892 MilliJy      104
 > comps
 > IMAGR2 11:08:12 VISDFT: Begin DFT component subtraction
 > IMAGR2 11:08:12 I Polarization model processed
 > IMAGR2 11:08:13 Field   1 min = -778.9 MicroJy,max =  773.9 MicroJy
 > IMAGR2 11:08:13 Loading field   1 to TV from -1.289E-03 to  3.015E-03
 > IMAGR2 11:08:27 TVDBOX: BOX( 1) =     2.00    1.00  512.00  512.00
 > IMAGR2 11:08:30 Turning off DOTV option - use TELL to turn on
 > IMAGR2 11:08:30 BGC Clean: using   15 cell beam + residuals >   293.17
 > MicroJy
 > IMAGR2 11:08:30    19984 Residual map points loaded
 > IMAGR2 11:08:30 SOMETHING IS GOING WRONG - ABANDON CLEAN
 > IMAGR2 11:08:30 Reached minimum algorithm flux = -577.870 MicroJy iter=
 > 205
 > IMAGR2 11:08:30 Total Cleaned flux density     =    6.302 MilliJy      205
 > comps

Note the TVDBOX message above.  Then 20000 pixels get cleaned in the
minor cycle rather than 100 or so.  The net Clean flux dropped a lot
which is what caused the ABANDON message.  Did you mean to Clean the
full area?

If so it is probably okay = otherwise you might want to repeat
things.

Eric Greisen