[daip] (no subject)
R. Craig Walker
cwalker at nrao.edu
Mon Jan 7 18:38:48 EST 2008
Here are some details for my apcal problem that I talked to you (Amy)
about today.
First I ran vlbacala with dofit=1 and all went normally.
Output began with lines like
localh> APCAL1: DFTRC:BR RCP Initial Trec set to = 49.39
localh> APCAL1: DFTRC:BR LCP Initial Trec set to = 80.64
Then I got the usual solution lines.
localh> APCAL1: Writing SN table 2
localh> APCAL1: BR RCP 0/16h 2m Trec (K): 68.84 Zen. opac.: .063
localh> APCAL1: BR LCP 0/16h 2m Trec (K): 81.46 Zen. opac.: .101
I wanted to use forced TRECVR for MK and still get an opacity from that
despite bad weather that made a sec z fit useless. I also wanted to do
normal fits for the other stations. I deleted the SN and CL tables from
the first run, then did:
tget apcal
DOFIT=1,1,1,1,1,0,1,1,1
TRECVR(11)=60; TRECVR(12)= 70
MK is antenna 6.
The Initial Trec lines are now missing and there is a single line (not one
per station) stating that the initial Tau0 is set to 0.1.
Then for all station except MK, I get the following from the fits:
localh> APCAL1: Writing SN table 3
localh> APCAL1: BR RCP 0/16h 2m Trec (K): .00 Zen. opac.: .201
localh> APCAL1: BR LCP 0/16h 2m Trec (K): .00 Zen. opac.: .201
For MK, I get no mention whatsoever. But it does appear in the plot file
with results that look about like what I got later running stand-alone.
So it looks like it treated MK ok, if silently, but messed up on the other
antennas.
Looking more at the plots, I think it used TRECVR=0.0 for all the other
antennas and did not allow that to change in the fit. And for some
reason,
got 0.201 for the opacity in all cases. The model line in the Tsys vs Sec
Z follows such a curve.
Here are the inputs.
>inp
AIPS 1: APCAL Task to generate an amplitude calibration SN table
AIPS 1: Adverbs Values Comments
AIPS 1: ----------------------------------------------------------------
AIPS 1: INNAME 'BW088PC' Input UV file name (name)
AIPS 1: INCLASS 'MULTI' Input UV file name (class)
AIPS 1: INSEQ 1 Input UV file name (seq. #)
AIPS 1: INDISK 4 Input UV file disk unit #
AIPS 1:
AIPS 1: ANTENNAS *all 0 Antennas to calibrate
AIPS 1: SUBARRAY 0 Subarray 0 -> 1
AIPS 1: STOKES ' ' Stokes type ('R','L',' ')
AIPS 1: BIF 0 Start IF number
AIPS 1: EIF 0 End IF number
AIPS 1: FREQID -1 Freq. ID to calibrate
AIPS 1: SOURCES *all ' ' Sources to calibrate
AIPS 1: TIMERANG *all 0 Time range to calibrate
AIPS 1:
AIPS 1: TYVER 1 TY table version number.
AIPS 1: GCVER 1 GC table version number.
AIPS 1: SNVER 0 Output SN table version.
AIPS 1: 0 -> highest unique
AIPS 1: OPCODE 'GRID' Opacity correction opcode:
AIPS 1: ' ' -> no opacity corr.
AIPS 1: 'GRID','GRDR','OPAC','OPCR',
AIPS 1: 'LESQ' -> fit for opacity
AIPS 1: APARM *all 0 Calibration parameters:
AIPS 1: (1) -> B factor (def. 1.0)
AIPS 1: Opacity parameters:
AIPS 1: (2) -> ZA limit (deg)
AIPS 1: (3) >=0 use spillover
AIPS 1: correction
AIPS 1: (4) > 0 use all FQ in
AIPS 1: opac. soln.
AIPS 1: (5) Limit to sec(z) plotted
AIPS 1: SOLINT 0 Solution interval (min)
AIPS 1: 0 -> use all data
AIPS 1: INVERS 1 WX table version number.
AIPS 1: 0 -> use text file
AIPS 1: INFILE *all ' ' Text file with weather info.
AIPS 1: Must be ' ' if using WX
AIPS 1: table (INVERS).
AIPS 1: TRECVR 0 0 Receiver temperatures (K);
AIPS 1: 0 0 in (R,L) pairs.
AIPS 1: 0 0 0 0
AIPS 1: 0 0 60 70
AIPS 1: *rest 0
AIPS 1: TAU0 *all 0 Zenith opacities (per ant.)
AIPS 1: DOFIT 1 1 Fit type (per ant.)
AIPS 1: 1 1 -1 -> no opacity correction
AIPS 1: 1 0 0 -> apply TRECVR only
AIPS 1: 1 1 1 -> solve for Trec using
AIPS 1: 1 *rest 0 fitting method spec.
AIPS 1: in OPCODE; then apply
AIPS 1: PRTLEV 0 Print level
AIPS 1: DOTV -1 TV selection (<0 -> PL file)
AIPS 1: LTYPE 3 Plot labeling
AIPS 1: GRCHAN 0 TV Graphics channel to use
I then tried running MK separately. First issue: is TRECVR indexed by AN
table number or ANTENNA number when ANTENNA is set? The help file is
clear about DOFIT, but not TRECVR.
First I reran the other antennas. That worked ok.
Then I set antenna=6,0 and trecvr=60,70, and dofit=0. It says it did not
do an opacity fit and the initial tau0 was set to 0.1, but otherwise is
mum.
Looking at the plot, it did give an opacity that might be reasonable. It
looks weird because it increases with time while the Tsys stayed fairly
constant. But that would not be too crazy if the opacity is a zenith
version, so it is dividing by a sec(z). So maybe that is about what was
wanted.
Now I'll combine the solutions in CLCAL and hope for the best.
Based on what I've seen so far, I think I'll try putting in TRECVR
values for all antennas and see if it gets a decent fit. Clearly this
message is documenting what I'm doing on the fly.
It worked! So, when you mix different DOFITs, you have to specify TRECVR
for all antennas. That's not too friendly, and I suggest a change. Or at
least a clear warning in the documentation. The best situation would be
if what it does to each antenna is independent of what it is told to do
for other antennas. I don't think we need to worry about the case where
the user really wants Trec=0 unless I'm missing something.
Cheers,
Craig
---------------------------------------------------------------------
R. Craig Walker Array Operations Center
cwalker at nrao.edu National Radio Astronomy Observatory
Phone 505 835 7247 P. O. Box O
Fax 505 835 7027 Socorro NM 87801 USA
--------------------------------------------------------------web
More information about the Daip
mailing list