shoste wrote: ↑Sun Aug 25, 2024 11:39 am
Dear Robin, Thanks for your kind reply and offer. As by magic the proprietary plate solver in Sharpcap works really well on 60 seconds exposures with my f/12 Takahashi Mewlon. Problem solved.
BUT, another issue came up. In measuring eclipse variables (e.g. a sequence 200 60 sec exposures), exposures taken after midnight automatically go into a new folder and are renumbered from 0h0m0s onwards. That means that the light curve is being split in two parts and rejoining them is not straightforward (apart for copying them to the directory of the day before, but then the hours are not consecutive any more...). Would it be possible to save sequences of files simply using Julian date and time (eg as a string with 5 digits after the comma) for the filename? That would put any sequence automatically in perfect chronological order. What about using the JDaverage time from the FITS header. Thanks for considering this. Do you have a better suggestion?
PS: I know that you can append yesterday's date {Date:Y} but that still renames the new files beginning with 0-0-0, and thus loosing the continuity again. Actually, sorting these would then place them in the wrong order.
I you are the USA and even better in Arizona USA, UTC is perfect 00:00 (UTC) hits us at 17:00 local time so the entire night or day is on the same UTC day. It makes finding series of lights so much easier and flats are easier to assign as well. In Arizona no DST and so we are always 7 hours off - perfect. Now as you move east this is not such an elegant solution New York 20:30 starts the new night and for Robin well not even a thought.
Now I do not like it when I ask for a solution, and someone tells me I am wrong for even asking which I am not doing! This is the first time I have heard of a JD solution and it sounds like some kind of derivation of JD could solve all your date problems no matter which time zone you observe from.
Sequence START date, Index number, time - this way the date stays the same as you wanted. The index number would keep them in order regardless of where midnight lands in the sequence.
Example
{Date:Y}_{Index:3}_{Time}
That way index 001 will be sorted before 002, etc. regardless of the time recorded.
kaymann wrote: ↑Sun Aug 25, 2024 2:25 pm
shoste you got me thinking - how would this work:
Sequence START date, Index number, time - this way the date stays the same as you wanted. The index number would keep them in order regardless of where midnight lands in the sequence.
Example
{Date:Y}_{Index:3}_{Time}
That way index 001 will be sorted before 002, etc. regardless of the time recorded.
Il give that solution a try.
@Robin, in the mean time your plate solving routine works a treat on a rather small field of my f/12 Mewlon with a tiny 7,5x5 mm sensor. Congratulations!
However, although the platesolve tools are all available at the beginning of a long series of 60 seconds exposures, they seem to wither away and end up mostly greyed after some time (one hour?). Except plate solve (solve only). Even then, the pushbutton for "plate solve and recenter" still operates correctly. Odd.
Your help is greatfully acknowledged. The solution to the "perpetually sortable" filenames problem was simple and elegant. In the Sharpcap settings, I tick:
include UTC, sortable dates and winjupos compatible
and that's it. I can even restart the computer and the fits files will still neatly follow each other. That I would have to join two sets of files, if the series extend over midnight, is then a minor hassle. Still, the ability to add Julian dates to the filenames remains an interesting option to consider.
Yours truly,
Serge
filenames 15s.PNG
You do not have the required permissions to view the files attached to this post.
