[Aavso-photometry] FITS and observation time

[Jeff Hopkins]

Hello Arne,

I did more testing with the DSI Pro and AutoStar to better understand 
how to get more precise timing information.

Observation Time Test
# Exp	Exp Time	Start	 Stop	    Info Box	 FITS Header
    10	  15 sec	05:40	08:25	10 Images 163 sec	Time 05:56
						10 images
 16 sec from end of first frame			150 sec


    20	  30 sec	12:50	23:18	20 Images 627 sec	Time 13:21
						20 Images
 31 sec from end of first frame			600 sec

    10	  30 sec	27:00	32:30	10 Images 328 sec	Time 27:46
						10 Images
 46 sec from end of first frame			300 sec


    20	  30 sec	34:00	44:29	20 Images 627 sec	Time 34:32
						20 Images
 32 sec from end of first frame			600 sec

Note: Start and Stop times are mm:ss.

For single exposures
Exposure		Info Box
15.0 sec		17 sec
11.3 sec		13 sec
8.1 sec		10 sec
5.7 sec		7 sec
4.0 sec		6 sec
2.8 sec		4 sec
2.0 sec		3 sec
1.4 sec		3 sec
1.0 sec		2 sec
0.7071 sec	2 sec
0.5 sec		2 sec
0.3536 sec	1 or 2 sec
Same for all faster times

Times are +/- 1 second

It seems the actual times can only be known to 1 or 2 seconds and 
times can vary I think due to processing activities of the computer. 
For stacked images it appears the best time estimate would be to take 
the time reported in the FITS header, subtract the individual 
exposure time then add one half the total exposure time reported in 
the FIS Header.

For example 20 - 30 second stacked exposures. The FITS Header time is 
hh:13:21 (note the hours are not used here to simplify things). The 
Total FITS Header exposure time is 600 seconds. The time of 
mid-observation would be approximately

hh:13:21 (780 seconds) - 30 + 600/2 = 750 + 300 = 1050 seconds or hh:17:30

I'd be very interested in any comments or suggestions. I have sent a 
request to Meade to allow more control over the time reported and to 
report in hh:mm:ss.ss instead of just hh:mm:ss.

Jeff



At 09:42 -0700 07/26/2007, arne wrote:
>Jeff Hopkins wrote:
>>  Normally I observe long period eclipsing binaries where precise time
>>  to the second or millisecond is not a factor. Recently I have started
>>  a mentor project using the fast eclipsing binary sz Herculis. I am
>>  using AutoStar and found the FITS time indication is off. It seems to
>>  be the time the file is saved. I started a one-shot 15 second
>>  exposure at 05:02:30. The end time should have been 05:02:45 or
>>  close. The FITS Header indicated 05:02:57.
>>
>>  If the difference is fairly constant, it could be compensated for.
>>  However, things get more complex when images are stacked and during
>>  the stacking when some images are not stacked due to poor quality.
>>
>>  How has this been handled by the AAVSO?
>>
>The AAVSO assumes that the submitted time is for the midpoint of
>the exposure.  Calculating this correctly is left up to the observer.
>
>Not quite sure how you would get 05:02:57 from your above example.
>Does Autostar take 12 seconds to download the image?  Certainly
>the written time cannot reflect the end-time of writing an image
>since the header gets written first.
>
>Most software packages modify the header time when stacking
>images.  There was an extensive discussion about this a couple
>of years ago on this list, but I'd presume that each software
>package writes things differently today.
>Arne
>_______________________________________________

-- 
Jeff Hopkins
HPO SOFT
Counting Photons
http://www.hposoft.com/Astro/astro.html
Hopkins Phoenix Observatory
7812 West Clayton Drive
Phoenix, Arizona 85033-2439 U.S.A.
(623)849-5889
(623) 247-1190 (Fax)
www.hposoft.com