[daip] Re: New CLCOR

[Leonia Kogan]

Mark,

Thank you very much for the good news about a good result at the first 
practical using of the new opcode='atmo' of CLCOR.

Now about the problem you met at CLCOR.

Did you meet it with the new opcode=ATMO?

I'll try to reproduce the problem here and fix it

Leonia






-----------------------------------------------------------------

Leonia,

    Both Andreas Brunthaler and I have been testing the new
version of CLCOR that you coded.  Andreas (in Bonn, Germany)
has had success and finds it does improve images noticeably.

I've been trying to test sign conventions and units.  I
obtained a test data set that Tony Beasley made some years
ago.  He processed a couple of hours of S/X data on two
sources (one was BL Lac) in different ways: once with
the normal correlator software and once with the atmospheric
model removed.

I FRINGed the test data and ran the multi-band delays into
my fitting program.  It gave me vertical atmospheric delays
exactly as expected (~230 x exp(-Elevation/8500m) cm).

So, next I tried to "fix up" the test data that had
zero atmospheres.  When I ran CLCOR with SOUR='', it
complained about no data for source #1.  (Indeed, there
is no data for that source, since I UVCOPed the data
and only retained source #2 = BL Lac).  However, it 
did **not** correct any entries in the CL table.  If I
specify SOUR='BLLAC', then it does change some CL table
entries.  Can this be fixed?

Thanks,
Mark

PS: I'll let you know how the full test worked shortly.