Date: Fri, 20 Jul 2012 09:38:32 -0400 (EDT)
From: Jun-Hui Zhao 
To: Berengere Parise 
Subject: Re: Miriad on Ubuntu for SMA data reduction



On Fri, 20 Jul 2012, Berengere Parise wrote:

> Dear Jui-Hui,
>
> Thanks for the new binaries. I was able to install them and run miriad.
> I could read the data file without problem.
>
> I have however the same problem as I had in June for the previous
> dataset: when using smamfcal, I get again a fatal error linked to the
> "edge" keyword.
> Last time, this could be solved by getting rid of the first two spectral
> windows
> that had only one data point.
> So I tried this with the 100822 dataset (s13 and s15 are empty, so I
> suspect
> they are causing the problem?).

     That was an embarrassment in keeping the original SMA correlator
configuration in a miriad file from smalod when involving a high-resolution
spectral mode with some empty chunks. As you have noticed, the smaloded
file conflicts with the edge function. For the first two empty windows,
your 2011 was simply using uvaver with select for the data
windows to get rid of the edge problem. The function 'select' only
handles consecutive windows. Here is a way to detour the problem
for your 2010 data assuming under c_shell and file name
of "100822_rx0.usb.tsys" after tsys correction (smafix) for the usb data.
Add the following lines in you c-shell script after smafix:
>>>>>>>>>>>>>>>>in a c_shell script <<<<<<<<<<<
....
set fname=100822_rx0.usb
\rm -r $fname.tsys_1 $fname.tsys_2 $fname.tsys_3
uvaver vis=$fname.tsys out=$fname.tsys_1
select='window(1,2,3,4,5,6,7,8,9,10,11,12)'
uvaver vis=$fname.tsys out=$fname.tsys_2 select='window(14)'
uvaver vis=$fname.tsys out=$fname.tsys_3
select='window(16,17,18,19,20,21,22,23,24)'
\rm -r $fname.tsys
smachunkglue vis=$fname.tsys out=$fname.tsys nfiles=3
uvindex vis=$fname.tsys
....
>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<
    Then, the rest of data processing should be similar to that of
your 2011 observation.
    Please let me know if you have further problems.
    Jun-Hui


>
> I tried the following, but get the following error:
>
> uvaver vis=data100822_rx0.lsb out=100822_rx0.lsb
> select='window(1,2,3,4,5,6,7,8,9,10,11,12,14,16,17,18,19,20,21,22,23,24)'
> UvAver: version 1.0 22-Jan-10
> ### Fatal Error [uvaver]:  Unsupported window selection clause, in
> UVREAD(init)
>
> This command worked fine and still works fine with the previous dataset
> I was
> calibrating, so I am a bit puzzled it does not work with the 100822 dataset?
>
> Cheers,
>
> Berengere
>
>
>
>