[daip] UV modelling / UVSUB

[Eric Greisen]

Hans-Rainer Kloeckner wrote:
> Dear Eric,
> 
> I hope you are fine and I also  hope you can help me, please. I'm 
> running into a problem I do not fully understand .
> 
> I try to model a visibility set for a GMRT dataset using UVSUB.
> 
> Now this is what I have done. 
> 
> 1. I have replaced all the real and imaginary values of the visibilities 
> to zero (doing all channels).
> 2. than I have add ONE point source at the phase centre to it, by using 
> UVSUB (7.33473 Jy, DFT,factor -1 ).
> 
> So far so good. Now if I have a look into the visibilities e.g. 
> amplitude versus time,  some of the  
> baselines show a rather strange amplitude behaviour always at the end of 
> the last scan (e.g. 1-29, 1-30). 
> It seems that the flux shouting up. Now the source is above 45 elevation 
> so not sure what causes this jump in amplitude.
> 
> Could that be related to some smoothing etc. ?
> 
> I have attached 2 files:
> 
> Amplitude versus time plot
> JJJ_MODEL_AMP_I.ps_pages file 
> 
> Model visibilities for antenna 1 
> JJJ.FITS
> 
> I hope you might have an idea.
> 
> cheers and thanks for hopefully considering to have a look into it.

I faked up an image with 1 CC and ran UVSUB with OPCODE 'MODL' and got 
what one would hope rather than any extraneous points.  Your wording 
above does not tell me quite what you did but I think I have replicated 
the main point (dropping real data, making 1 point source) and all is as 
expected.  Where your result comes from I cannot say.

Eric Greisen