[daip] aips imagr

[Eric Greisen]

Cao Hongmin wrote:
> 
> Hi Eric
> 
> i am ambiguous with the delay search window and the rate search window. 
> whether the following calculation is proper:
> Suppose the solution time (not the correlation integration/average time) 
> for the fringe-fitting is 2min, the default rate search window will be 
> [-(1/120)*2*1000 mHz, (1/120)*2*1000 mHz ]; if 16MHz/IF (choose to solve 
> single band delay), then the default delay search window would be 
> [-1/32*1000 ns, 1/32*1000 ns].
> 
> I misunderstand the meaning of a sentence in the help file : "...BOXFILE 
> must be used for fields numbered > 64..."
> then if there are less fields (than 64) that want to image, "Boxfile" 
> can still be used.
> For the primary beam correction, we just use a Gaussian model to correct 
> the primary beam attenuation (EVN antennas haven't proper primary beam 
> measurement, at the moment). if there is a task, we just set some 
> parameters, such as, "nfield¡°, "weight", "channel average", " time 
> average", " signal to noise cutoff", etc, then it can help us to search 
> the weak source. if get one, it just shift the phase centre to the 
> position of the source and averaged the data and then split it out. that 
> will be awesome :). Multiple phase centre techniques need a source 
> coordinate list in advance, but some times the coordinates are prob no 
> enough accuracy.

The rate search window is based on the averaging time in the data set 
not SOLINT.  The delay window is based on the frequency spacing in the 
data set not including CHINC so the frequency spacing data presented to 
the solver.  The windows used are recorded by FRING in the history file.

We do not have a magic task to do your science.  FLATN will make a 
single image from your multiple facets.  You may then search it, or 
smoothed versions of it for maxima.  You may then SPLIT with coordinate 
shifts.

Eric Greisen