[daip] CCEDT and CALIB problems
Amy Mioduszewski
amiodusz at nrao.edu
Thu Aug 28 12:49:21 EDT 2008
HI Andy,
Obviously this is something Eric did this year. Just FYI, another new feature
which effects solution intervals, is SPLIT now creates NX tables. CALIB will
not let the solution interval be longer than a scan in NX table. I don't think
that probably has anything to do with your current problem.
O.K., can you copy out a small time range of your data, make sure it shows this
problem and put it somewhere I can fetch it.
Thanks,
Amy
Andy Biggs wrote:
> Hi Amy. I've just run CALIB on data at my old institute (this time on a
> Linux box) where I have access to 'tst' and 'new' versions of AIPS. The
> results are different:
>
> 31DEC07 31DEC08
>
> Good: 5910 6056
> Failed: 326 376
>
> I've also plotted the amplitude solutions for each (attached) and in the
> 31DEC07 version the anomalous points between scans are absent.
>
> I hope this helps.
>
> Andy
>
> Amy Mioduszewski wrote:
>> Hi Andy,
>>
>> Do you have access to an older version of AIPS? Eric changed how
>> SOLINT was calculated in CALIB in July. If you do have access to an
>> older version try CALIB with that and tell me if you get the same thing.
>>
>> Amy
>>
>> Andy Biggs wrote:
>>> Hi Amy. Thanks for your help. Personally, I still think that CCEDT
>>> should output an empty CC table if no clean components 'make the
>>> grade' so to speak; if that change could be made it would be a great
>>> help. Thanks for clarifying what IMAGR does with regard to this.
>>>
>>> However, I have another problem, this time with CALIB. I'm finding
>>> solutions *in between* scans of the source that I'm calibrating with.
>>> I have a multi-source file with alternate scans of my phase
>>> calibrator (~3-min duration) and my target (~25 min duration). With
>>> solution intervals of 2-3 min on my target things are fine. However,
>>> if the solution interval goes to 4 min I then start seeing solutions
>>> *between* scans. Not all scans are affected, just a couple usually,
>>> but the effect is noticeable in SNPLT in two ways:
>>>
>>> 1) The gaps which are obvious between the other target scans are not
>>> visible
>>>
>>> 2) The solutions between the scans are obviously 'not right'
>>>
>>> I've attached a postscript file that shows what I'm talking about.
>>> There are 4 and a bit scans shown. Each scan receives 7 solutions,
>>> but between two of the scans there are additional, discrepant,
>>> solutions. Looking at the solutions using LISTR OPTY='GAIN' confirms
>>> that the solution timestamp corresponds to the middle-ish of the
>>> phase-calibrator scan.
>>>
>>> I seem to recall from a midnight job email recently that some work
>>> was done on how CALIB decides on positions of solution intervals;
>>> maybe this is related? Let me reiterate that everything is fine until
>>> I increase the solution interval to 4 min. It might be significant
>>> that 4 min is longer than the gaps between target scans (3.23 min).
>>>
>>> I've also tried this on a file where I SPLATted the target source out
>>> of the multi-source file i.e. there is only data from the target. I
>>> see the same behaviour. Looking at the NX table, all scans have the
>>> same gap between them so why only some would be affected I don't
>>> know. Here are my inputs:
>>>
>>> AIPS 2: CALIB: Task to determine calibration for data.
>>> AIPS 2: Adverbs Values Comments
>>> AIPS 2: ----------------------------------------------------------------
>>> AIPS 2: Input uv data.
>>> AIPS 2: INNAME 'TEST' UV file name (name)
>>> AIPS 2: INCLASS 'SPLAT' UV file name (class)
>>> AIPS 2: INSEQ 1 UV file name (seq. #)
>>> AIPS 2: INDISK 1 UV file disk drive #
>>> AIPS 2: Data selection (multisource):
>>> AIPS 2: CALSOUR 'TEST' Calibrator sources
>>> AIPS 2: *rest ' '
>>> AIPS 2: QUAL -1 Calibrator qualifier -1=>all
>>> AIPS 2: CALCODE ' ' Calibrator code ' '=>all
>>> AIPS 2: SELBAND -1 Bandwidth to select (kHz)
>>> AIPS 2: SELFREQ -1 Frequency to select (MHz)
>>> AIPS 2: FREQID -1 Freq. ID to select.
>>> AIPS 2: TIMERANG *all 0 Time range to use.
>>> AIPS 2: BCHAN 1 Lowest channel number 0=>all
>>> AIPS 2: ECHAN 7 Highest channel number
>>> AIPS 2: ANTENNAS *all 0 Antennas to select. 0=all
>>> AIPS 2: DOFIT *all 0 Subset of ANTENNAS list for
>>> AIPS 2: which solns are desired.
>>> AIPS 2: ANTUSE *all 0 Mean gain is calculated
>>> AIPS 2: (CPARM(2)>0) using only the
>>> AIPS 2: listed antennas. See explain.
>>> AIPS 2: SUBARRAY 0 Subarray, 0=>all
>>> AIPS 2: UVRANGE 0 0 Range of uv distance for full
>>> AIPS 2: weight
>>> AIPS 2: WTUV 0 Weight outside UVRANGE 0=0.
>>> AIPS 2: WEIGHTIT 0 Modify data weights function
>>> AIPS 2: Cal. info for input:
>>> AIPS 2: DOCALIB 2 > 0 calibrate data & weights
>>> AIPS 2: > 99 do NOT calibrate weights
>>> AIPS 2: GAINUSE 5 CL table to apply.
>>> AIPS 2: DOPOL -1 If >0 correct polarization.
>>> AIPS 2: BLVER -1 BL table to apply.
>>> AIPS 2: FLAGVER 0 Flag table version
>>> AIPS 2: DOBAND -1 If >0 apply bandpass cal.
>>> AIPS 2: Method used depends on value
>>> AIPS 2: of DOBAND (see HELP file).
>>> AIPS 2: BPVER -1 Bandpass table version
>>> AIPS 2: SMOOTH *all 0 Smoothing function. See
>>> AIPS 2: HELP SMOOTH for details.
>>> AIPS 2:
>>> AIPS 2: CLEAN map. See HELP.
>>> AIPS 2: IN2NAME 'TEST' Cleaned map name (name)
>>> AIPS 2: IN2CLASS 'ICL001' Cleaned map name (class)
>>> AIPS 2: IN2SEQ 1 Cleaned map name (seq. #)
>>> AIPS 2: IN2DISK 1 Cleaned map disk unit #
>>> AIPS 2: INVERS 0 CC file version #.
>>> AIPS 2: NCOMP *all 0 # comps to use for model.
>>> AIPS 2: 1 value per field
>>> AIPS 2: FLUX 0 Lowest CC component used.
>>> AIPS 2: NMAPS 59 No. Clean map files
>>> AIPS 2: CMETHOD ' ' Modeling method:
>>> AIPS 2: 'DFT','GRID',' '
>>> AIPS 2: CMODEL ' ' Model type: 'COMP','IMAG'
>>> AIPS 2: SMODEL *all 0 Source model, 1=flux,2=x,3=y
>>> AIPS 2: ** press RETURN for more, enter Q or next line to quit print **
>>> AIPS 2: See HELP SMODEL for models.
>>> AIPS 2:
>>> AIPS 2: Output uv data file.
>>> AIPS 2: OUTNAME ' ' UV file name (name)
>>> AIPS 2: OUTCLASS ' ' UV file name (class)
>>> AIPS 2: OUTSEQ 0 UV file name (seq. #)
>>> AIPS 2: OUTDISK 1 UV file disk drive #
>>> AIPS 2:
>>> AIPS 2: Solution control adverbs:
>>> AIPS 2: REFANT 10 Reference antenna
>>> AIPS 2: SOLINT 4 Solution interval (min)
>>> AIPS 2: SOLSUB 0 Solution subinterval
>>> AIPS 2: SOLMIN 0 Min solution interval
>>> AIPS 2: APARM *all 0 General parameters
>>> AIPS 2: 1=min. no. antennas
>>> AIPS 2: 2 > 0 => data divided
>>> AIPS 2: 3 > 0 => avg. RR,LL
>>> AIPS 2: 5 > 0 => avg. IFs.
>>> AIPS 2: 6=print level, 1=good,
>>> AIPS 2: 2 closure, 3 SNR
>>> AIPS 2: 7=SNR cutoff (0=>5)
>>> AIPS 2: 8=max. ant. # (no AN)
>>> AIPS 2: 9 > 0 => pass failed soln
>>> AIPS 2: 10 < 99 cal output weights
>>> AIPS 2: Phase-amplitude Parameters:
>>> AIPS 2: DOFLAG 0 Flag on closure error?
>>> AIPS 2: SOLTYPE ' ' Soln type,' ','L1','GCON',
>>> AIPS 2: 'R', 'L1R', 'GCOR'
>>> AIPS 2: SOLMODE 'A&P' Soln. mode: 'A&P','P','P!A',
>>> AIPS 2: 'GCON',
>>> AIPS 2: SOLCON 0 Gain constraint factor.
>>> AIPS 2: MINAMPER 0 Amplitude closure error
>>> AIPS 2: regarded as excessive in %
>>> AIPS 2: MINPHSER 0 Phase closure error regarded
>>> AIPS 2: as excessive in degrees
>>> AIPS 2: CPARM *all 0 Phase-amp. parameters
>>> AIPS 2: 1 = Min el for gain
>>> AIPS 2: normalization (deg)
>>> AIPS 2: 2 >0 => normalize gain
>>> AIPS 2: 3 avg. amp. closure err
>>> AIPS 2: 4 avg. ph. closure err
>>> AIPS 2: 5 >0 => scalar average
>>> AIPS 2: 6 limit clipping in robust
>>> AIPS 2: 7 limit display of closure
>>> AIPS 2: errors
>>> AIPS 2: SNVER 0 Output SN table, 0=>new table
>>> AIPS 2: ANTWT *all 0 Ant. weights (0=>1.0)
>>> AIPS 2: GAINERR *all 0 Std. Dev. of antenna gains.
>>> AIPS 2: BADDISK *all 0 Disk no. not to use for
>>> AIPS 2: scratch files.
>>>
>>> Thanks,
>>>
>>> Andy
>>>
>>> --
>>>
>>> Andy Biggs
>>> ESO, ALMA Regional Centre
>>> Karl-Schwarzschild-Strasse 2
>>> D-85748 Garching, Germany
>>>
>>> tel. +49-89-32006471
>>> fax. +49-89-32006898
>>>
>>>
>>> ------------------------------------------------------------------------
>>>
>>> _______________________________________________
>>> Daip mailing list
>>> Daip at listmgr.cv.nrao.edu
>>> http://listmgr.cv.nrao.edu/mailman/listinfo/daip
>
More information about the Daip
mailing list