[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Trying to do it



Tom

1)
1d is the recommended standard in one of the FITS documents.  It is directly
decodable by the CFITSIO library and hopefully others, too.  The time is UT, not
local.

DATE-OBS = '2001-03-04T02:08:09.3'

2) Where is 2)?

4)
EXPTIME is required by some IRAF routines. The format is
EXPTIME  = 15.0   /  Exposure time in seconds

If you put EXPOSURE with the same format, it will probably not hurt.
 
> 
> 5)   USHUT    =     '16:34:45.0'     /UT of shutter open
> 

I do not know the purpose of USHUT.  Maybe it is required by some specific
software.  But the information contained in DATE-OBS should be exactly the same:
The time when the shutter opened.

> 
> 6)  GAIN    '7.5'                     / electrons per ADU
> 
When you use numerical values, you certainly don't need '' or "", also for
logical values such as T and F.  You put them around strings. So GAIN = 7.5.

> 
> 9)  WCS keywords have been requested.  I have no idea how to compute these
> values.
> 
When you take images, you put in only approximations.  Any decent processing
program will compute "real" (=better approximation) values on the basis of
catalog matching and insert them in.  But:

CRPIX1 = (image width)/2
CRPIX2 = (image height)/2

CDELT1, CDELT2 are roughly 2.08e-3 degree (7.5 arc seconds) for MARK IV.  I
suggest to put the real values into a configuration file which is read at the
start-up of the image aqusition program.  Note that they may (must) be negative
if the celestial coordinates decrease with the increasing image coordinates.

CROTA1 and CROTA2 are ideally 0.  I assume you will eventually orient the camera
along the RA-DEC.

CRVAL1 and CRVAL2 are calculated from your internal counters of motor steps after
you calibrate the telescope.  They represent celestial coordinates at the
CRPIX1,CRPIX2 image coordinates and are expressed in decimal degrees, i.e. for
ra=13:00:00 CRVAL1=195.

> 10)  From the same request:
> 
> DATE-OBS = '13/07/97'
> HDR_REV = '2001-Mar-05'
> 
> What form do we want to use for dates?  I am a confirmed "Little Endian" so
> I like to write dates as 20010305.234 for 2001-Mar-05.  What does fits expect?

Use the same form for dates as in DATE-OBS in 1).


					Jure