Changes between Version 1 and Version 2 of Notes/FirstNightReduction


Ignore:
Timestamp:
04/02/09 15:48:56 (15 years ago)
Author:
JamieStevens
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • Notes/FirstNightReduction

    v1 v2  
    1717The files we are using are:
    1818{{{
    19 4.5G 2009-04-02 11:14 2009-04-01_1407.C999
    20 4.7G 2009-04-02 11:15 2009-04-01_1543.C999
    21 2.8G 2009-04-02 11:17 2009-04-01_1723.C999
    22 1.9G 2009-04-02 11:17 2009-04-01_1822.C999
    23 3.1G 2009-04-02 11:18 2009-04-01_1938.C999
     194.5G 2009-04-01_1407.C999
     204.7G 2009-04-01_1543.C999
     212.8G 2009-04-01_1723.C999
     221.9G 2009-04-01_1822.C999
     233.1G 2009-04-01_1938.C999
    2424}}}
    2525for a total of 17G.
     
    3131  in       = *.C999
    3232  out      = firstnight_1934.uv
    33   ifsel    = 1,2
     33  ifsel    =  
    3434  restfreq = 
    35   options  =  
     35  options  = noif
    3636  nfiles   = 
    3737  nscans   = 
    3838}}}
    3939
     40We use options=noif since there are actually 4 IFs in each CABB file: IFs 1 & 2 are the 2 GHz 2048 channel
     41wide-bands, while IFs 3 and 4 are the 4096 channel areas that the zoom bands will use. Since the IFs have
     42different numbers of channels, ATLOD will fail if you ask it to map the IFs to the same channel axis. After
     43using options=noif you may select the different IFs using the select=window options.
     44
     45Let's look at the data using uvplt (this takes quite a while with all the data!):
     46{{{
     47uvplt% inp
     48  Task:   uvplt
     49  vis      = firstnight_1934.uv/
     50  line     = 
     51  select   = -auto,window(1),pol(xx,yy)
     52  stokes   = 
     53  axis     = time,amp
     54  xrange   = 
     55  yrange   = 
     56  average  = 
     57  hann     = 
     58  inc      = 
     59  options  = nobase
     60  subtitle = 
     61  device   = firstnight_firstlook.png/png
     62  nxy      = 
     63  size     = 
     64  log      = 
     65  comment  = 
     66}}}
     67
     68[[Image(firstnight_firstlook.png)]]
     69
     70This image shows us that something weird happened - we didn't observe on April 2! A quick
     71uvindex shows the problem:
     72{{{
     73uvindex% inp
     74  Task:   uvindex
     75  vis      = firstnight_1934.uv/
     76  interval = 
     77  log      = 
     78  options  = 
     79}}}
     80{{{
     8109APR01:16:58:33.2 aa1              n  6      2049        0      1  318505
     8209APR01:16:59:34.0 aa2              n  6      2049        0      1  321097
     8309APR01:17:00:34.0 aa3              n  6      2049        0      1  323689
     8409APR02:17:01:27.4 aa3              n  6      2049        0      1  326009
     8509APR02:17:02:03.2 bb1              n  6      2049        0      1  326225
     8609APR02:17:03:04.0 bb2              n  6      2049        0      1  328817
     8709APR02:17:04:04.0 bb3              n  6      2049        0      1  331409
     8809APR01:17:05:38.2 cc1              n  6      2049        0      1  334001
     8909APR01:17:06:39.0 cc2              n  6      2049        0      1  336593
     9009APR01:17:07:39.0 cc3              n  6      2049        0      1  339185
     91}}}
     92
     93A little quirk has meant that the timestamp is wrong for a few scans. Since the quirk may have
     94also corrupted the data, we take no chances and flag it out with uvflag:
     95{{{
     96uvflag% inp
     97  Task:   uvflag
     98  vis      = firstnight_1934.uv/
     99  select   = time(09APR02:00:00:00)
     100  line     = 
     101  edge     = 
     102  flagval  = flag
     103  options  = 
     104  log      = 
     105}}}
     106
     107And looking at the data again shows us the following plot.
     108
     109[[Image(firstnight_aftertimeflag.png)]]
     110
     111So we can see that we are observing different sources and that the data looks as one would
     112expect it to. Now let's just look at 1934-638, as we will use it as our primary calibrator.
     113{{{
     114uvplt% inp
     115  Task:   uvplt
     116  vis      = firstnight_1934.uv/
     117  line     = 
     118  select   = -auto,window(1),pol(xx,yy),source(aa1)
     119  stokes   = 
     120  axis     = time,amp
     121  xrange   = 
     122  yrange   = 
     123  average  = 
     124  hann     = 
     125  inc      = 
     126  options  = nobase
     127  subtitle = 
     128  device   = 1/xs
     129  nxy      = 
     130  size     = 
     131  log      = 
     132  comment  = 
     133}}}
     134Here we use the source name aa1 for the mosaic pointing on 1934-638 with no offset.
     135
     136[[Image(firstnight_1934_beforecal.png)]]
     137
     138And we can plot phase too:
     139
     140[[Image(firstnight_1934p_beforecal.png)]]
     141
     142And the bandpass:
     143{{{
     144uvspec% inp
     145  Task:   uvspec
     146  vis      = firstnight_1934.uv/
     147  select   = -auto,window(1),pol(xx,yy),source(aa1)
     148  line     = 
     149  stokes   = 
     150  interval = 10
     151  hann     = 
     152  offset   = 
     153  options  = nobase,avall
     154  axis     = freq,amp
     155  yrange   = 
     156  device   = firstnight_1934b_beforecal.png/png
     157  nxy      = 3,4
     158  log      = 
     159}}}
     160
     161[[Image(firstnight_1934b_beforecal.png)]]
     162
     163Now we calibrate:
     164{{{
     165mfcal% inp
     166  Task:   mfcal
     167  vis      = firstnight_1934.uv/
     168  line     = 
     169  stokes   = 
     170  edge     = 
     171  select   = window(1),source(aa1)
     172  flux     = 
     173  refant   = 
     174  minants  = 
     175  interval = 0.1
     176  options  = 
     177  tol      = 
     178}}}
     179