Can't see data in some windows

[Maria Jose Maureira]

Hi,

I was working on calibrating the data and for some windows I can't see anything. I was doing a uvcat to select the window that I wanted 

 uvcat vis=$vis select="win($win_wide),-auto,-source(noise)" out=$out

and I got this error:

### Warning [uvcat]:  Variable corr has zero or negative size, in UVPUTVR

uvcat(9143) malloc: *** error for object 0x7fafaa404fb8: incorrect checksum for freed object - object was probably modified after being freed.

*** set a breakpoint in malloc_error_break to debug

Abort

I went to check the data and I can't

smauvplt device=/xs select='win(2),purpose(s),-auto' vis=../track2/raw/c1184.2D_90B1bSan.6.miriad options=all

SmaUvPlt: version 1.8 1-dec-09

Will plot flagged and unflagged data

File # 1  = ../track2/raw/c1184.2D_90B1bSan.6.miriad

88.6232 GHz

n=0 start=319 step=1 width=1 nchan=319

### Fatal Error [smauvplt]:  Illegal channel range specified, in UVREAD

In some cases I can see the gain calibrator and/or flux cal and in others windows I can't see anything. Does anybody know what could be going on?

Thanks a lot!

Maria 

[Manuel Fernández]

Maria Jose,

Creo que el problema que te esta dando esta posiblemente relacionado con la version de Miriad que estas usando... anteriormetne he visto esos malloc errors vinculados con la version de Miriad y las nuevas computadoras de 64 bits. 

Son esos los datos de la propuesta que metimos para los FH cores? Si yo estoy en la propuesta puedo echarles un ojo el lunes, para ver si hay algun problema con ellos, si quieres.

Saludos,

manuel

*******************************************************
Manuel Fernández López
University of Illinois at Urbana-Champaign

manf...@illinois.edu

manf...@gmail.com

--
You received this message because you are subscribed to the Google Groups "miriad" group.
To unsubscribe from this group and stop receiving emails from it, send an email to miriad+un...@googlegroups.com.
To post to this group, send email to mir...@googlegroups.com.
Visit this group at http://groups.google.com/group/miriad.
For more options, visit https://groups.google.com/groups/opt_out.

[Peter Teuben]

can you try a

uvindex vis=$vis

and see if you have sections of time where a window dropped out?
Probably best, if not too often,
to flag those time ranges, or work around it with some creative uvcat'ing?

peter

[Maria Jose Maureira]

Hi!

I did an uvindex and I got this for the track that it is bad

   Time        Source        Antennas Spectral Wideband  Freq  Record

                    Name                  Channels Channels Config   No.

14FEB18:23:20:46.5 NOISE            23      4544       16      1       1

14FEB18:23:39:57.0 URANUS           23      4544       16      1     361

14FEB18:23:57:24.5 NOISE            23      4544       16      1    1561

14FEB18:23:58:11.0 3C84             23      4544       16      1    1681

14FEB19:00:02:13.0 B1B-NS           23      4544       16      1    2401

14FEB19:00:09:02.0 B1B-NS           23      3906       16      2    3961

14FEB19:00:09:32.0 B1B-NS           23      4544       16      1    4081

14FEB19:00:13:56.5 B1B-NS           23      3906       16      3    5041

14FEB19:00:14:26.5 B1B-NS           23      4544       16      1    5161

14FEB19:00:16:55.5 NOISE            23      4544       16      1    5761

14FEB19:00:18:08.5 3C84             23      4544       16      1    5881

.

.

.

and this for the track that is good

14JAN29:00:04:27.0 NOISE            23      4544       16      1       1

14JAN29:00:13:56.0 URANUS           23      4544       16      1     241

14JAN29:00:27:42.5 NOISE            23      4544       16      1    1441

14JAN29:00:28:29.5 3C84             23      4544       16      1    1561

14JAN29:00:32:21.5 B1B-NS           23      4544       16      1    2281

14JAN29:00:47:05.5 NOISE            23      4544       16      1    5641

14JAN29:00:48:02.0 3C84             23      4544       16      1    5761

14JAN29:00:51:57.5 B1B-NS           23      4544       16      1    6481

14JAN29:01:06:41.5 NOISE            23      4544       16      1    9841

14JAN29:01:07:42.0 3C84             23      4544       16      1    9961

14JAN29:01:11:43.0 B1B-NS           23      4544       16      1   10681

14JAN29:01:26:27.5 NOISE            23      4544       16      1   14041

.

.

.

So, it seems that there are some windows that disappeared in between pointings, but I don't understand why. Also what is that freq configuration column? both tracks have the same correlator conf. I have another source that have the same problem...

Best,

Maria Jose

[Chat Hull]

Yes, the best thing to do is to flag the time ranges with fewer-than-normal channels.  I had to do this with some of the TADPOL data...generally flagging one second before until one second after worked.  Then you might have to rewrite the data with options=unflagged.

Chat

Chat Hull
Astronomy Department
UC Berkeley

LinkedIn: linkedin.com/in/chathull

Web: astro.berkeley.edu/~chat

[Maria Jose Maureira]

That was exactly what I did!

Thanks so much for the help,

Maria Jose