BF072A reduction notes:

BF072A on 14 March 2004 = day 71

This run, like some others, suffers from the general problem that
the R/L phases drift with time.  There might be enough SNR 
to correct this if the IFs are averaged in CALIB.  Maybe we have
to make three passes through CALIB?

Yes, this works.  
findstrong identifies strong engough sources and then TABPUT identifies
them as "E"xtra calibrators

EVPA calibration data:

           5                        8                  22
J0854    48 mJy @ -40         91 mJy @ -18        156 mJy @ 7.4  April 12
3C279   534 mJy @ -172       516 mJy @ 146        540 mJy @ 138  April 12
J1310    22 mJy @ -4          52 mJy @ -15        106 mJy @ -10  April 12

Array name = 'VLBA    '   includes antennas:
           1:BR         2:FD         3:HN
           4:KP         5:LA         6:MK
           7:NL         8:OV         9:PT
          10:SC
refant=2

NL has high Tsys (raining?) for first couple hours

5 GHz
Step 1:  VIPSPIPE

3C279  timer 0 7 44 0 0 7 46 20 for fringe-fit
calsour '3c279','oq208','j1310+3220','J0854+2006',''
interp 'ambg'

AIPS 1: 'SUMMARY OF THE CL-TABLES PRODUCED FOR THIS FREQUENCY-ID:'
AIPS 1: 'CL-1 : ANTENNAS AND TIMES PRESENT'
AIPS 1: 'CL-2 : SAMPLER CORRECTIONS ADDED'
AIPS 1: 'CL-3 : GAIN CORRECTIONS ADDED'
AIPS 1: 'CL-4 : PARALLACTIC ANGLE CORRECTION ADDED'
AIPS 1: 'CL-5 : (MANUAL?) P-CAL ADDED'
AIPS 1: 'CL-6 : IONOSPHERIC CORRECTION ADDED'
AIPS 1: 'CL-7 : TOTAL CALIBRATION TABLE'
         CL-8   R-L delay calibration
AIPS 1: ' '
AIPS 1: 'THESE PLOTS HAVE BEEN PRODUCED FOR THIS FREQUENCY-ID:'
AIPS 1: ' '
AIPS 1: '1-10   ANTENNAS AND TIMES PRESENT'
AIPS 1: '11-20  SAMPLER CORRECTIONS IN SN'
AIPS 1: '21-30 GAIN CORRECTIONS IN SN'
AIPS 1: '51-51 IONOSPHERIC DELAY CORRECTION'
AIPS 1: '52-55 BANDPASS CORRECTIONS'
AIPS 1: '56-85 FRINGE-FIT SOLUTIONS IN SN'
AIPS 1: '86-95 SNR OF FRINGED SOLUTIONS'
AIPS 1: '96-135 TOTAL CALIBRATION TABLE'
AIPS 1: ' '

Step 2: making models with Difmap
doitall

Step 3: polarization calibration
makestep3
set inputs
step3

Step 4: final imaging
# set the bottom contour level on the following line in Jy/beam
# set x = `grep $n table2.txt | nawk '{print ($8/100 * $6)/1000}'`
set x = 0.0006

doitall

15 GHz:

timer 0 7 42 0 0 7 44 20  for manual pcal

EVPA calibration:


Step 4:
doitall

Step 5:

run and go step5, being carefull to clear the VIPSDATA area of *.TBU1

EVPA calibration
first step of lining up the IFs happens automatically then:

doitall on the new *.TBU1 created by the procedure

use:
loadall; step5pall  to load in the new images and look at the EVPAs

VLA measurements (converted to degrees) from March 13
J0854* 4860     38  -1     3C279 577 -84   J1310 27  -8
J0854* 15000    42  17     3C279 501  63   J1310 46  -7  ; extrapolated with sm

*J0854 from data on March 27.  

VLBA observables: (use IF 3 = 4.992 GHz)
J0854* 4860     37  90     3C279 518  -2   J1310 25  71
   corr            -91               -86            -79  ; use -86 degrees
                  (use IF 4 = 15.366 GHz)
J0854* 15000    88 -82     3C279 520 -77   J1310 68  30  
   corr            -81               -40            -43  ; use -42

tget CLCOR; gainver 4; gainuse 5; go CLCOR 
tget SPLIT; gainuse 5; go SPLIT

use doitall on the new *.TBU1
loadall; step5pall  to load in the final images and polarization maps

3C279 C4 at pixxy 139 121 is 80 degrees after all corrections at 5 GHz
3C279 C4 at pixxy 280 240 is 74 degrees after all corrections at 15 GHz

Analysis:

08815606  
08835114 is polarized at 5, 15
08845198
1543+517 is polarized at 5, 15
15675806
15765401 is polarized at 5, 15

make Faraday RM and other plots

loadpair; step5init; loadpair  to pull in first 5 then 15 GHz paired sets
at fixed resolution

step5pall; to make the POLC and POLA maps

step5rm ; to make the rm images

15 GHz has baseline sensitivity limit of ~30 mJy.  Several sources
had peaks around this level suggesting that self-calibration was
biasing the peak up.  

Re-imaged all 15 GHz by pairing the IFs.  This produced more reliable
flux densities, although sources < 40 mJy should be viewed with 
caution.  sources < 25 mJy are probably not detected (3).  If we go
to 256 Mbps then we should be able to detect sources > 18 mJy.
Here are the non-detections:

0851+580 (false peak = 22 mJy )
08975587 (false peak = 17 mJy)
15675806 (false peak = 23 mJy)  possible detection in visibility data

What's in the directories
step6:
15GHz/         The 4 IF data after all calibration
15GHz_allif/   Averaged all IFs together after all calibration **NOT USED**
15GHz_fix/     Fixed resolution for paired IFs
15GHz_pairif/  Averaged pairs of IFs (1+2 and 3+4) after calibration
                  currently polarization is from 15GHz_fix
                      total inensity made here and copied to 15GHz_fix
5GHz/          The 4 IF data after all calibration
5GHz_fix/      Fixed resolution for paired IFs in pol'n
big5/          Big fields (150 x 150 mas)

results:
The matched resolution 5 and 15 GHz images, + SPIX + ROTMES