[daip] (no subject)

[Eric Greisen]

novae hmc wrote:
> 
> when doing the fringe-fitting in aips, we adopt 2 steps:
> one step is that when running  the task "fring", just use a strong 
> source and 2min or less , then do clcal to produce a new cl table;
> the other step is that choose all the calibrators and 2 min of solution 
> time when running the task "fring", and then do another clcal.
> it is correct or not to use different *reference antenna* in the two 
> steps? how about the following task "bpass" calibration after the 
> fringe-fitting?
************ Please put an informative subject in any e-mail.  Many 
filters delete all mail with no subject and I often delete them without 
reading.  This time I did not.  ******************

I asked a VLBI expert to comment and he wrote:


I'm presuming that he is doing the two steps sequentially.  Either alone 
could get him to a very similar final result unless some of the 
calibrators are so week that you need to narrow the window significantly 
to get good detections.  In fact, for high frequency work (where the 
ionosphere isn't pushing the delay all over the place), I often find it 
adequate to just to step one as the only fringe fit.

The main question seems to be whether it is ok to switch calibrators. If 
you are not doing polarization calibration, I don't see any particular 
reason why not, although I would not normally do such a thing.  In 
principle it should work but I suppose you might get into a bit of 
trouble with rapid fluctuations.  Also if one or the other calibrator is 
not there all the time, you might have issues.  With polarization 
calibration, you need to be careful that the left/right phase and delay 
differences don't change.  The easiest way to do that is to never make 
adjustments to one of the antennas, the reference antenna.  If you 
switch reference antennas, you might get away with it, but you would 
need to think and test very carefully what you are doing.  Two major 
steps of polarization calibration are to remove any left/right delay 
difference (RLDLY) of the reference antenna and to calibrate the 
left/right phase difference (EVPA calibration).  If those are changing 
as might happen when you change reference antennas, the calibration 
could go badly astray.

For the same reasons as above, I use the same reference antenna for the 
bandpass calibration.  Again, it probably doesn't matter for total power 
imaging, but the phase part of the bandpass is capable of messing with 
the left/right phase difference.

Eric Greisen